{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.2013742701672161, "global_step": 10140, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 9.92063492063492e-07, "loss": 10.8421, "theoretical_loss": 20.81281780154715, "tokens_seen": 65536 }, { "epoch": 0.0, "learning_rate": 1.984126984126984e-06, "loss": 10.794, "theoretical_loss": 17.566201104328645, "tokens_seen": 131072 }, { "epoch": 0.0, "learning_rate": 2.9761904761904763e-06, "loss": 10.7002, "theoretical_loss": 15.939477092836569, "tokens_seen": 196608 }, { "epoch": 0.0, "learning_rate": 3.968253968253968e-06, "loss": 10.6201, "theoretical_loss": 14.89231675598857, "tokens_seen": 262144 }, { "epoch": 0.0, "learning_rate": 4.96031746031746e-06, "loss": 10.1305, "theoretical_loss": 14.136216937762974, "tokens_seen": 327680 }, { "epoch": 0.0, "learning_rate": 5.9523809523809525e-06, "loss": 10.084, "theoretical_loss": 13.552561472550224, "tokens_seen": 393216 }, { "epoch": 0.0, "learning_rate": 6.944444444444444e-06, "loss": 9.8364, "theoretical_loss": 13.08180900140119, "tokens_seen": 458752 }, { "epoch": 0.0, "learning_rate": 7.936507936507936e-06, "loss": 9.897, "theoretical_loss": 12.690129625483323, "tokens_seen": 524288 }, { "epoch": 0.0, "learning_rate": 8.928571428571428e-06, "loss": 9.5044, "theoretical_loss": 12.356592463873625, "tokens_seen": 589824 }, { "epoch": 0.0, "learning_rate": 9.92063492063492e-06, "loss": 9.6321, "theoretical_loss": 12.067412607035077, "tokens_seen": 655360 }, { "epoch": 0.0, "learning_rate": 1.0912698412698412e-05, "loss": 9.4371, "theoretical_loss": 11.813066231101676, "tokens_seen": 720896 }, { "epoch": 0.0, "learning_rate": 1.1904761904761905e-05, "loss": 9.6176, "theoretical_loss": 11.586719208706729, "tokens_seen": 786432 }, { "epoch": 0.0, "learning_rate": 1.2896825396825396e-05, "loss": 9.199, "theoretical_loss": 11.383314140186787, "tokens_seen": 851968 }, { "epoch": 0.0, "learning_rate": 1.3888888888888888e-05, "loss": 8.8935, "theoretical_loss": 11.199011702111871, "tokens_seen": 917504 }, { "epoch": 0.0, "learning_rate": 1.4880952380952381e-05, "loss": 9.3252, "theoretical_loss": 11.030833917977912, "tokens_seen": 983040 }, { "epoch": 0.0, "learning_rate": 1.5873015873015872e-05, "loss": 8.6232, "theoretical_loss": 10.87642808645695, "tokens_seen": 1048576 }, { "epoch": 0.0, "learning_rate": 1.6865079365079364e-05, "loss": 9.0944, "theoretical_loss": 10.733905740062724, "tokens_seen": 1114112 }, { "epoch": 0.0, "learning_rate": 1.7857142857142855e-05, "loss": 8.899, "theoretical_loss": 10.60172987623028, "tokens_seen": 1179648 }, { "epoch": 0.0, "learning_rate": 1.884920634920635e-05, "loss": 8.9962, "theoretical_loss": 10.478634172356642, "tokens_seen": 1245184 }, { "epoch": 0.0, "learning_rate": 1.984126984126984e-05, "loss": 9.3635, "theoretical_loss": 10.36356394376333, "tokens_seen": 1310720 }, { "epoch": 0.0, "learning_rate": 2.0833333333333333e-05, "loss": 8.7939, "theoretical_loss": 10.255632220896747, "tokens_seen": 1376256 }, { "epoch": 0.0, "learning_rate": 2.1825396825396824e-05, "loss": 8.8135, "theoretical_loss": 10.15408655327002, "tokens_seen": 1441792 }, { "epoch": 0.0, "learning_rate": 2.2817460317460315e-05, "loss": 8.5781, "theoretical_loss": 10.058283561732598, "tokens_seen": 1507328 }, { "epoch": 0.0, "learning_rate": 2.380952380952381e-05, "loss": 9.316, "theoretical_loss": 9.967669178840278, "tokens_seen": 1572864 }, { "epoch": 0.0, "objective/train/docs_used": 13112, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 8.743459701538086, "objective/train/theoretical_loss": 9.881763126393109, "objective/train/tokens_used": 22098400, "theoretical_loss": 9.881763126393109, "tokens_seen": 1638400 }, { "epoch": 0.0, "learning_rate": 2.48015873015873e-05, "loss": 8.8669, "theoretical_loss": 9.881763126393109, "tokens_seen": 1638400 }, { "epoch": 0.0, "learning_rate": 2.5793650793650793e-05, "loss": 8.868, "theoretical_loss": 9.80014659154056, "tokens_seen": 1703936 }, { "epoch": 0.0, "learning_rate": 2.6785714285714284e-05, "loss": 8.3477, "theoretical_loss": 9.722452346907446, "tokens_seen": 1769472 }, { "epoch": 0.0, "learning_rate": 2.7777777777777776e-05, "loss": 8.5573, "theoretical_loss": 9.648356759081546, "tokens_seen": 1835008 }, { "epoch": 0.0, "learning_rate": 2.876984126984127e-05, "loss": 8.7737, "theoretical_loss": 9.577573271145639, "tokens_seen": 1900544 }, { "epoch": 0.0, "learning_rate": 2.9761904761904762e-05, "loss": 8.9664, "theoretical_loss": 9.509847046764852, "tokens_seen": 1966080 }, { "epoch": 0.0, "learning_rate": 3.075396825396825e-05, "loss": 7.8002, "theoretical_loss": 9.444950537631936, "tokens_seen": 2031616 }, { "epoch": 0.0, "learning_rate": 3.1746031746031745e-05, "loss": 8.6073, "theoretical_loss": 9.382679790910457, "tokens_seen": 2097152 }, { "epoch": 0.0, "learning_rate": 3.273809523809524e-05, "loss": 8.7208, "theoretical_loss": 9.32285135423398, "tokens_seen": 2162688 }, { "epoch": 0.0, "learning_rate": 3.373015873015873e-05, "loss": 8.7212, "theoretical_loss": 9.265299666660276, "tokens_seen": 2228224 }, { "epoch": 0.0, "learning_rate": 3.472222222222222e-05, "loss": 8.7873, "theoretical_loss": 9.209874847444755, "tokens_seen": 2293760 }, { "epoch": 0.0, "learning_rate": 3.571428571428571e-05, "loss": 8.4388, "theoretical_loss": 9.156440812508292, "tokens_seen": 2359296 }, { "epoch": 0.0, "learning_rate": 3.670634920634921e-05, "loss": 8.1114, "theoretical_loss": 9.10487366241335, "tokens_seen": 2424832 }, { "epoch": 0.0, "learning_rate": 3.76984126984127e-05, "loss": 8.2518, "theoretical_loss": 9.055060296533734, "tokens_seen": 2490368 }, { "epoch": 0.0, "learning_rate": 3.8690476190476195e-05, "loss": 8.2561, "theoretical_loss": 9.006897216643829, "tokens_seen": 2555904 }, { "epoch": 0.0, "learning_rate": 3.968253968253968e-05, "loss": 8.1331, "theoretical_loss": 8.960289489909357, "tokens_seen": 2621440 }, { "epoch": 0.0, "learning_rate": 4.067460317460318e-05, "loss": 8.0787, "theoretical_loss": 8.915149846640611, "tokens_seen": 2686976 }, { "epoch": 0.0, "learning_rate": 4.1666666666666665e-05, "loss": 8.3421, "theoretical_loss": 8.871397892478225, "tokens_seen": 2752512 }, { "epoch": 0.0, "learning_rate": 4.265873015873016e-05, "loss": 8.0099, "theoretical_loss": 8.828959418153499, "tokens_seen": 2818048 }, { "epoch": 0.0, "learning_rate": 4.365079365079365e-05, "loss": 8.2887, "theoretical_loss": 8.787765792778412, "tokens_seen": 2883584 }, { "epoch": 0.0, "learning_rate": 4.464285714285714e-05, "loss": 8.5295, "theoretical_loss": 8.747753428911455, "tokens_seen": 2949120 }, { "epoch": 0.0, "learning_rate": 4.563492063492063e-05, "loss": 8.2913, "theoretical_loss": 8.708863309520833, "tokens_seen": 3014656 }, { "epoch": 0.0, "learning_rate": 4.6626984126984126e-05, "loss": 8.319, "theoretical_loss": 8.671040568508847, "tokens_seen": 3080192 }, { "epoch": 0.0, "learning_rate": 4.761904761904762e-05, "loss": 7.8599, "theoretical_loss": 8.634234117735474, "tokens_seen": 3145728 }, { "epoch": 0.0, "learning_rate": 4.8611111111111115e-05, "loss": 7.7079, "theoretical_loss": 8.598396314536323, "tokens_seen": 3211264 }, { "epoch": 0.0, "objective/train/docs_used": 13755, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 7.561188220977783, "objective/train/theoretical_loss": 8.563482664611069, "objective/train/tokens_used": 23736800, "theoretical_loss": 8.563482664611069, "tokens_seen": 3276800 }, { "epoch": 0.0, "learning_rate": 4.96031746031746e-05, "loss": 7.8004, "theoretical_loss": 8.563482664611069, "tokens_seen": 3276800 }, { "epoch": 0.0, "learning_rate": 5.05952380952381e-05, "loss": 8.0798, "theoretical_loss": 8.529451555895115, "tokens_seen": 3342336 }, { "epoch": 0.0, "learning_rate": 5.1587301587301586e-05, "loss": 7.9158, "theoretical_loss": 8.496264019646002, "tokens_seen": 3407872 }, { "epoch": 0.0, "learning_rate": 5.257936507936508e-05, "loss": 7.5747, "theoretical_loss": 8.463883515497187, "tokens_seen": 3473408 }, { "epoch": 0.0, "learning_rate": 5.357142857142857e-05, "loss": 7.8708, "theoretical_loss": 8.432275737672779, "tokens_seen": 3538944 }, { "epoch": 0.0, "learning_rate": 5.4563492063492063e-05, "loss": 7.7568, "theoretical_loss": 8.401408439930716, "tokens_seen": 3604480 }, { "epoch": 0.0, "learning_rate": 5.555555555555555e-05, "loss": 7.86, "theoretical_loss": 8.371251277120209, "tokens_seen": 3670016 }, { "epoch": 0.0, "learning_rate": 5.6547619047619046e-05, "loss": 7.5876, "theoretical_loss": 8.341775661511075, "tokens_seen": 3735552 }, { "epoch": 0.0, "learning_rate": 5.753968253968254e-05, "loss": 7.0476, "theoretical_loss": 8.31295463228533, "tokens_seen": 3801088 }, { "epoch": 0.0, "learning_rate": 5.8531746031746036e-05, "loss": 7.6287, "theoretical_loss": 8.284762736781182, "tokens_seen": 3866624 }, { "epoch": 0.0, "learning_rate": 5.9523809523809524e-05, "loss": 7.8486, "theoretical_loss": 8.257175922251864, "tokens_seen": 3932160 }, { "epoch": 0.0, "learning_rate": 6.051587301587302e-05, "loss": 7.7722, "theoretical_loss": 8.230171437050114, "tokens_seen": 3997696 }, { "epoch": 0.0, "learning_rate": 6.15079365079365e-05, "loss": 7.7011, "theoretical_loss": 8.20372774027797, "tokens_seen": 4063232 }, { "epoch": 0.0, "learning_rate": 6.25e-05, "loss": 7.8601, "theoretical_loss": 8.177824419053046, "tokens_seen": 4128768 }, { "epoch": 0.0, "learning_rate": 6.349206349206349e-05, "loss": 7.4565, "theoretical_loss": 8.152442112639616, "tokens_seen": 4194304 }, { "epoch": 0.0, "learning_rate": 6.448412698412699e-05, "loss": 7.3253, "theoretical_loss": 8.1275624427775, "tokens_seen": 4259840 }, { "epoch": 0.0, "learning_rate": 6.547619047619048e-05, "loss": 7.1366, "theoretical_loss": 8.10316794961571, "tokens_seen": 4325376 }, { "epoch": 0.0, "learning_rate": 6.646825396825397e-05, "loss": 7.4872, "theoretical_loss": 8.07924203272264, "tokens_seen": 4390912 }, { "epoch": 0.0, "learning_rate": 6.746031746031745e-05, "loss": 7.5183, "theoretical_loss": 8.055768896701416, "tokens_seen": 4456448 }, { "epoch": 0.0, "learning_rate": 6.845238095238096e-05, "loss": 7.0588, "theoretical_loss": 8.032733500989007, "tokens_seen": 4521984 }, { "epoch": 0.0, "learning_rate": 6.944444444444444e-05, "loss": 7.3674, "theoretical_loss": 8.010121513461836, "tokens_seen": 4587520 }, { "epoch": 0.0, "learning_rate": 7.043650793650793e-05, "loss": 7.0203, "theoretical_loss": 7.987919267509379, "tokens_seen": 4653056 }, { "epoch": 0.0, "learning_rate": 7.142857142857142e-05, "loss": 7.0155, "theoretical_loss": 7.966113722271801, "tokens_seen": 4718592 }, { "epoch": 0.0, "learning_rate": 7.242063492063492e-05, "loss": 7.2468, "theoretical_loss": 7.944692425767988, "tokens_seen": 4784128 }, { "epoch": 0.0, "learning_rate": 7.341269841269842e-05, "loss": 7.022, "theoretical_loss": 7.9236434806675184, "tokens_seen": 4849664 }, { "epoch": 0.0, "objective/train/docs_used": 15099, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 6.9753007888793945, "objective/train/theoretical_loss": 7.902955512484067, "objective/train/tokens_used": 25375200, "theoretical_loss": 7.902955512484067, "tokens_seen": 4915200 }, { "epoch": 0.0, "learning_rate": 7.440476190476191e-05, "loss": 7.1984, "theoretical_loss": 7.902955512484067, "tokens_seen": 4915200 }, { "epoch": 0.0, "learning_rate": 7.53968253968254e-05, "loss": 7.0639, "theoretical_loss": 7.882617639989203, "tokens_seen": 4980736 }, { "epoch": 0.0, "learning_rate": 7.63888888888889e-05, "loss": 6.8249, "theoretical_loss": 7.862619447664628, "tokens_seen": 5046272 }, { "epoch": 0.0, "learning_rate": 7.738095238095239e-05, "loss": 6.9557, "theoretical_loss": 7.842950960027937, "tokens_seen": 5111808 }, { "epoch": 0.0, "learning_rate": 7.837301587301588e-05, "loss": 7.1772, "theoretical_loss": 7.823602617682313, "tokens_seen": 5177344 }, { "epoch": 0.0, "learning_rate": 7.936507936507937e-05, "loss": 6.7487, "theoretical_loss": 7.804565254954165, "tokens_seen": 5242880 }, { "epoch": 0.0, "learning_rate": 8.035714285714287e-05, "loss": 6.7244, "theoretical_loss": 7.7858300789950725, "tokens_seen": 5308416 }, { "epoch": 0.0, "learning_rate": 8.134920634920635e-05, "loss": 7.2012, "theoretical_loss": 7.767388650235364, "tokens_seen": 5373952 }, { "epoch": 0.0, "learning_rate": 8.234126984126984e-05, "loss": 6.7945, "theoretical_loss": 7.749232864086619, "tokens_seen": 5439488 }, { "epoch": 0.0, "learning_rate": 8.333333333333333e-05, "loss": 6.4889, "theoretical_loss": 7.731354933799318, "tokens_seen": 5505024 }, { "epoch": 0.0, "learning_rate": 8.432539682539683e-05, "loss": 7.0238, "theoretical_loss": 7.71374737438992, "tokens_seen": 5570560 }, { "epoch": 0.0, "learning_rate": 8.531746031746032e-05, "loss": 6.9503, "theoretical_loss": 7.696402987558934, "tokens_seen": 5636096 }, { "epoch": 0.0, "learning_rate": 8.630952380952381e-05, "loss": 6.8864, "theoretical_loss": 7.679314847528181, "tokens_seen": 5701632 }, { "epoch": 0.0, "learning_rate": 8.73015873015873e-05, "loss": 6.5425, "theoretical_loss": 7.662476287731328, "tokens_seen": 5767168 }, { "epoch": 0.0, "learning_rate": 8.82936507936508e-05, "loss": 6.9681, "theoretical_loss": 7.645880888297279, "tokens_seen": 5832704 }, { "epoch": 0.0, "learning_rate": 8.928571428571429e-05, "loss": 6.4495, "theoretical_loss": 7.629522464270861, "tokens_seen": 5898240 }, { "epoch": 0.0, "learning_rate": 9.027777777777777e-05, "loss": 6.5397, "theoretical_loss": 7.613395054519696, "tokens_seen": 5963776 }, { "epoch": 0.0, "learning_rate": 9.126984126984126e-05, "loss": 6.4287, "theoretical_loss": 7.59749291128028, "tokens_seen": 6029312 }, { "epoch": 0.0, "learning_rate": 9.226190476190476e-05, "loss": 6.5709, "theoretical_loss": 7.581810490299888, "tokens_seen": 6094848 }, { "epoch": 0.0, "learning_rate": 9.325396825396825e-05, "loss": 6.7154, "theoretical_loss": 7.5663424415343705, "tokens_seen": 6160384 }, { "epoch": 0.0, "learning_rate": 9.424603174603175e-05, "loss": 6.4567, "theoretical_loss": 7.551083600364949, "tokens_seen": 6225920 }, { "epoch": 0.0, "learning_rate": 9.523809523809524e-05, "loss": 6.351, "theoretical_loss": 7.536028979299919, "tokens_seen": 6291456 }, { "epoch": 0.0, "learning_rate": 9.623015873015874e-05, "loss": 6.7204, "theoretical_loss": 7.521173760129762, "tokens_seen": 6356992 }, { "epoch": 0.0, "learning_rate": 9.722222222222223e-05, "loss": 6.9272, "theoretical_loss": 7.506513286506497, "tokens_seen": 6422528 }, { "epoch": 0.0, "learning_rate": 9.821428571428572e-05, "loss": 6.3143, "theoretical_loss": 7.492043056920249, "tokens_seen": 6488064 }, { "epoch": 0.0, "objective/train/docs_used": 15777, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 6.954127788543701, "objective/train/theoretical_loss": 7.4777587180480305, "objective/train/tokens_used": 27013600, "theoretical_loss": 7.4777587180480305, "tokens_seen": 6553600 }, { "epoch": 0.0, "learning_rate": 9.92063492063492e-05, "loss": 6.6163, "theoretical_loss": 7.4777587180480305, "tokens_seen": 6553600 }, { "epoch": 0.0, "learning_rate": 0.00010019841269841271, "loss": 6.3178, "theoretical_loss": 7.463656058451462, "tokens_seen": 6619136 }, { "epoch": 0.0, "learning_rate": 0.0001011904761904762, "loss": 6.3722, "theoretical_loss": 7.449731002601916, "tokens_seen": 6684672 }, { "epoch": 0.0, "learning_rate": 0.00010218253968253968, "loss": 6.2211, "theoretical_loss": 7.435979605213019, "tokens_seen": 6750208 }, { "epoch": 0.0, "learning_rate": 0.00010317460317460317, "loss": 6.5174, "theoretical_loss": 7.422398045861905, "tokens_seen": 6815744 }, { "epoch": 0.0, "learning_rate": 0.00010416666666666667, "loss": 6.6538, "theoretical_loss": 7.408982623881875, "tokens_seen": 6881280 }, { "epoch": 0.0, "learning_rate": 0.00010515873015873016, "loss": 6.1797, "theoretical_loss": 7.395729753510345, "tokens_seen": 6946816 }, { "epoch": 0.0, "learning_rate": 0.00010615079365079365, "loss": 6.5865, "theoretical_loss": 7.3826359592770325, "tokens_seen": 7012352 }, { "epoch": 0.0, "learning_rate": 0.00010714285714285714, "loss": 6.5482, "theoretical_loss": 7.369697871618373, "tokens_seen": 7077888 }, { "epoch": 0.0, "learning_rate": 0.00010813492063492064, "loss": 6.7304, "theoretical_loss": 7.3569122227050885, "tokens_seen": 7143424 }, { "epoch": 0.0, "learning_rate": 0.00010912698412698413, "loss": 6.2506, "theoretical_loss": 7.3442758424706875, "tokens_seen": 7208960 }, { "epoch": 0.0, "learning_rate": 0.00011011904761904761, "loss": 6.1023, "theoretical_loss": 7.331785654829519, "tokens_seen": 7274496 }, { "epoch": 0.0, "learning_rate": 0.0001111111111111111, "loss": 6.6284, "theoretical_loss": 7.319438674073677, "tokens_seen": 7340032 }, { "epoch": 0.0, "learning_rate": 0.0001121031746031746, "loss": 6.3081, "theoretical_loss": 7.307232001438824, "tokens_seen": 7405568 }, { "epoch": 0.0, "learning_rate": 0.00011309523809523809, "loss": 6.299, "theoretical_loss": 7.295162821829564, "tokens_seen": 7471104 }, { "epoch": 0.0, "learning_rate": 0.00011408730158730158, "loss": 6.5472, "theoretical_loss": 7.283228400695652, "tokens_seen": 7536640 }, { "epoch": 0.0, "learning_rate": 0.00011507936507936508, "loss": 6.292, "theoretical_loss": 7.271426081050832, "tokens_seen": 7602176 }, { "epoch": 0.0, "learning_rate": 0.00011607142857142858, "loss": 6.2414, "theoretical_loss": 7.259753280626623, "tokens_seen": 7667712 }, { "epoch": 0.0, "learning_rate": 0.00011706349206349207, "loss": 6.3899, "theoretical_loss": 7.24820748915387, "tokens_seen": 7733248 }, { "epoch": 0.0, "learning_rate": 0.00011805555555555556, "loss": 6.3743, "theoretical_loss": 7.236786265765262, "tokens_seen": 7798784 }, { "epoch": 0.0, "learning_rate": 0.00011904761904761905, "loss": 5.9537, "theoretical_loss": 7.225487236512497, "tokens_seen": 7864320 }, { "epoch": 0.0, "learning_rate": 0.00012003968253968255, "loss": 6.0297, "theoretical_loss": 7.21430809199212, "tokens_seen": 7929856 }, { "epoch": 0.0, "learning_rate": 0.00012103174603174604, "loss": 6.5923, "theoretical_loss": 7.2032465850744005, "tokens_seen": 7995392 }, { "epoch": 0.0, "learning_rate": 0.00012202380952380953, "loss": 6.116, "theoretical_loss": 7.192300528730015, "tokens_seen": 8060928 }, { "epoch": 0.0, "learning_rate": 0.000123015873015873, "loss": 6.4398, "theoretical_loss": 7.1814677939495155, "tokens_seen": 8126464 }, { "epoch": 0.0, "objective/train/docs_used": 17166, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 6.351791858673096, "objective/train/theoretical_loss": 7.1707463077509646, "objective/train/tokens_used": 28652000, "theoretical_loss": 7.1707463077509646, "tokens_seen": 8192000 }, { "epoch": 0.0, "learning_rate": 0.0001240079365079365, "loss": 6.4479, "theoretical_loss": 7.1707463077509646, "tokens_seen": 8192000 }, { "epoch": 0.0, "learning_rate": 0.000125, "loss": 6.335, "theoretical_loss": 7.160134051271272, "tokens_seen": 8257536 }, { "epoch": 0.0, "learning_rate": 0.0001259920634920635, "loss": 6.1976, "theoretical_loss": 7.149629057937138, "tokens_seen": 8323072 }, { "epoch": 0.0, "learning_rate": 0.00012698412698412698, "loss": 6.4173, "theoretical_loss": 7.139229411711638, "tokens_seen": 8388608 }, { "epoch": 0.0, "learning_rate": 0.00012797619047619048, "loss": 6.0317, "theoretical_loss": 7.128933245412794, "tokens_seen": 8454144 }, { "epoch": 0.0, "learning_rate": 0.00012896825396825398, "loss": 6.0573, "theoretical_loss": 7.118738739100616, "tokens_seen": 8519680 }, { "epoch": 0.0, "learning_rate": 0.00012996031746031748, "loss": 6.0885, "theoretical_loss": 7.1086441185293445, "tokens_seen": 8585216 }, { "epoch": 0.0, "learning_rate": 0.00013095238095238096, "loss": 6.3385, "theoretical_loss": 7.09864765366177, "tokens_seen": 8650752 }, { "epoch": 0.0, "learning_rate": 0.00013194444444444446, "loss": 6.1813, "theoretical_loss": 7.088747657242693, "tokens_seen": 8716288 }, { "epoch": 0.0, "learning_rate": 0.00013293650793650793, "loss": 6.368, "theoretical_loss": 7.078942483428749, "tokens_seen": 8781824 }, { "epoch": 0.0, "learning_rate": 0.00013392857142857144, "loss": 5.9535, "theoretical_loss": 7.069230526471966, "tokens_seen": 8847360 }, { "epoch": 0.0, "learning_rate": 0.0001349206349206349, "loss": 6.0473, "theoretical_loss": 7.059610219454568, "tokens_seen": 8912896 }, { "epoch": 0.0, "learning_rate": 0.0001359126984126984, "loss": 6.0929, "theoretical_loss": 7.0500800330726685, "tokens_seen": 8978432 }, { "epoch": 0.0, "learning_rate": 0.0001369047619047619, "loss": 6.3183, "theoretical_loss": 7.040638474466625, "tokens_seen": 9043968 }, { "epoch": 0.0, "learning_rate": 0.00013789682539682541, "loss": 6.2457, "theoretical_loss": 7.031284086095933, "tokens_seen": 9109504 }, { "epoch": 0.0, "learning_rate": 0.0001388888888888889, "loss": 6.2984, "theoretical_loss": 7.022015444656678, "tokens_seen": 9175040 }, { "epoch": 0.0, "learning_rate": 0.0001398809523809524, "loss": 6.1863, "theoretical_loss": 7.012831160039609, "tokens_seen": 9240576 }, { "epoch": 0.0, "learning_rate": 0.00014087301587301586, "loss": 6.3604, "theoretical_loss": 7.003729874327071, "tokens_seen": 9306112 }, { "epoch": 0.0, "learning_rate": 0.00014186507936507937, "loss": 6.0677, "theoretical_loss": 6.994710260827057, "tokens_seen": 9371648 }, { "epoch": 0.0, "learning_rate": 0.00014285714285714284, "loss": 6.2727, "theoretical_loss": 6.98577102314278, "tokens_seen": 9437184 }, { "epoch": 0.0, "learning_rate": 0.00014384920634920634, "loss": 6.2095, "theoretical_loss": 6.976910894276189, "tokens_seen": 9502720 }, { "epoch": 0.0, "learning_rate": 0.00014484126984126984, "loss": 6.2549, "theoretical_loss": 6.968128635764015, "tokens_seen": 9568256 }, { "epoch": 0.0, "learning_rate": 0.00014583333333333335, "loss": 6.3308, "theoretical_loss": 6.959423036844894, "tokens_seen": 9633792 }, { "epoch": 0.0, "learning_rate": 0.00014682539682539685, "loss": 6.364, "theoretical_loss": 6.950792913656309, "tokens_seen": 9699328 }, { "epoch": 0.0, "learning_rate": 0.00014781746031746032, "loss": 6.0516, "theoretical_loss": 6.942237108460029, "tokens_seen": 9764864 }, { "epoch": 0.0, "objective/train/docs_used": 17861, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 6.117774486541748, "objective/train/theoretical_loss": 6.9337544888949, "objective/train/tokens_used": 30290400, "theoretical_loss": 6.9337544888949, "tokens_seen": 9830400 }, { "epoch": 0.0, "learning_rate": 0.00014880952380952382, "loss": 6.1082, "theoretical_loss": 6.9337544888949, "tokens_seen": 9830400 }, { "epoch": 0.0, "learning_rate": 0.0001498015873015873, "loss": 5.8009, "theoretical_loss": 6.925343947255817, "tokens_seen": 9895936 }, { "epoch": 0.0, "learning_rate": 0.0001507936507936508, "loss": 5.8948, "theoretical_loss": 6.917004399797798, "tokens_seen": 9961472 }, { "epoch": 0.0, "learning_rate": 0.00015178571428571427, "loss": 5.8529, "theoretical_loss": 6.908734786064147, "tokens_seen": 10027008 }, { "epoch": 0.0, "learning_rate": 0.0001527777777777778, "loss": 6.0551, "theoretical_loss": 6.900534068237688, "tokens_seen": 10092544 }, { "epoch": 0.0, "learning_rate": 0.00015376984126984128, "loss": 6.1428, "theoretical_loss": 6.89240123051416, "tokens_seen": 10158080 }, { "epoch": 0.0, "learning_rate": 0.00015476190476190478, "loss": 5.741, "theoretical_loss": 6.884335278496871, "tokens_seen": 10223616 }, { "epoch": 0.0, "learning_rate": 0.00015575396825396825, "loss": 5.8161, "theoretical_loss": 6.87633523861175, "tokens_seen": 10289152 }, { "epoch": 0.0, "learning_rate": 0.00015674603174603175, "loss": 6.1526, "theoretical_loss": 6.868400157541997, "tokens_seen": 10354688 }, { "epoch": 0.0, "learning_rate": 0.00015773809523809523, "loss": 5.9937, "theoretical_loss": 6.860529101681551, "tokens_seen": 10420224 }, { "epoch": 0.0, "learning_rate": 0.00015873015873015873, "loss": 6.0325, "theoretical_loss": 6.85272115660663, "tokens_seen": 10485760 }, { "epoch": 0.0, "learning_rate": 0.0001597222222222222, "loss": 5.938, "theoretical_loss": 6.844975426564642, "tokens_seen": 10551296 }, { "epoch": 0.0, "learning_rate": 0.00016071428571428573, "loss": 6.1597, "theoretical_loss": 6.8372910339797945, "tokens_seen": 10616832 }, { "epoch": 0.0, "learning_rate": 0.0001617063492063492, "loss": 5.92, "theoretical_loss": 6.829667118974749, "tokens_seen": 10682368 }, { "epoch": 0.0, "learning_rate": 0.0001626984126984127, "loss": 5.4308, "theoretical_loss": 6.8221028389077185, "tokens_seen": 10747904 }, { "epoch": 0.0, "learning_rate": 0.00016369047619047618, "loss": 6.2676, "theoretical_loss": 6.814597367924395, "tokens_seen": 10813440 }, { "epoch": 0.0, "learning_rate": 0.00016468253968253969, "loss": 6.0017, "theoretical_loss": 6.807149896524181, "tokens_seen": 10878976 }, { "epoch": 0.0, "learning_rate": 0.00016567460317460316, "loss": 6.1886, "theoretical_loss": 6.799759631140145, "tokens_seen": 10944512 }, { "epoch": 0.0, "learning_rate": 0.00016666666666666666, "loss": 5.7028, "theoretical_loss": 6.7924257937322245, "tokens_seen": 11010048 }, { "epoch": 0.0, "learning_rate": 0.00016765873015873016, "loss": 6.1481, "theoretical_loss": 6.785147621393148, "tokens_seen": 11075584 }, { "epoch": 0.0, "learning_rate": 0.00016865079365079366, "loss": 5.8603, "theoretical_loss": 6.777924365966638, "tokens_seen": 11141120 }, { "epoch": 0.0, "learning_rate": 0.00016964285714285717, "loss": 5.6239, "theoretical_loss": 6.770755293677423, "tokens_seen": 11206656 }, { "epoch": 0.0, "learning_rate": 0.00017063492063492064, "loss": 5.6818, "theoretical_loss": 6.763639684772625, "tokens_seen": 11272192 }, { "epoch": 0.0, "learning_rate": 0.00017162698412698414, "loss": 5.7517, "theoretical_loss": 6.756576833174123, "tokens_seen": 11337728 }, { "epoch": 0.0, "learning_rate": 0.00017261904761904762, "loss": 5.7609, "theoretical_loss": 6.749566046141486, "tokens_seen": 11403264 }, { "epoch": 0.0, "objective/train/docs_used": 18493, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.560274600982666, "objective/train/theoretical_loss": 6.7426066439450905, "objective/train/tokens_used": 31928800, "theoretical_loss": 6.7426066439450905, "tokens_seen": 11468800 }, { "epoch": 0.0, "learning_rate": 0.00017361111111111112, "loss": 5.9297, "theoretical_loss": 6.7426066439450905, "tokens_seen": 11468800 }, { "epoch": 0.0, "learning_rate": 0.0001746031746031746, "loss": 5.8221, "theoretical_loss": 6.735697959549075, "tokens_seen": 11534336 }, { "epoch": 0.0, "learning_rate": 0.0001755952380952381, "loss": 6.0875, "theoretical_loss": 6.728839338303761, "tokens_seen": 11599872 }, { "epoch": 0.0, "learning_rate": 0.0001765873015873016, "loss": 5.6213, "theoretical_loss": 6.722030137647226, "tokens_seen": 11665408 }, { "epoch": 0.0, "learning_rate": 0.0001775793650793651, "loss": 5.8156, "theoretical_loss": 6.715269726815689, "tokens_seen": 11730944 }, { "epoch": 0.0, "learning_rate": 0.00017857142857142857, "loss": 5.6855, "theoretical_loss": 6.7085574865624125, "tokens_seen": 11796480 }, { "epoch": 0.0, "learning_rate": 0.00017956349206349207, "loss": 6.0002, "theoretical_loss": 6.701892808884824, "tokens_seen": 11862016 }, { "epoch": 0.0, "learning_rate": 0.00018055555555555555, "loss": 5.4675, "theoretical_loss": 6.695275096759559, "tokens_seen": 11927552 }, { "epoch": 0.0, "learning_rate": 0.00018154761904761905, "loss": 5.6532, "theoretical_loss": 6.68870376388518, "tokens_seen": 11993088 }, { "epoch": 0.0, "learning_rate": 0.00018253968253968252, "loss": 5.6652, "theoretical_loss": 6.682178234432274, "tokens_seen": 12058624 }, { "epoch": 0.0, "learning_rate": 0.00018353174603174602, "loss": 5.7053, "theoretical_loss": 6.675697942800715, "tokens_seen": 12124160 }, { "epoch": 0.0, "learning_rate": 0.00018452380952380953, "loss": 5.638, "theoretical_loss": 6.669262333383815, "tokens_seen": 12189696 }, { "epoch": 0.0, "learning_rate": 0.00018551587301587303, "loss": 5.9672, "theoretical_loss": 6.662870860339158, "tokens_seen": 12255232 }, { "epoch": 0.0, "learning_rate": 0.0001865079365079365, "loss": 5.785, "theoretical_loss": 6.656522987365879, "tokens_seen": 12320768 }, { "epoch": 0.0, "learning_rate": 0.0001875, "loss": 5.7804, "theoretical_loss": 6.6502181874881705, "tokens_seen": 12386304 }, { "epoch": 0.0, "learning_rate": 0.0001884920634920635, "loss": 5.7629, "theoretical_loss": 6.643955942844831, "tokens_seen": 12451840 }, { "epoch": 0.0, "learning_rate": 0.00018948412698412698, "loss": 5.7272, "theoretical_loss": 6.637735744484626, "tokens_seen": 12517376 }, { "epoch": 0.0, "learning_rate": 0.00019047619047619048, "loss": 5.9396, "theoretical_loss": 6.631557092167304, "tokens_seen": 12582912 }, { "epoch": 0.0, "learning_rate": 0.00019146825396825398, "loss": 5.6727, "theoretical_loss": 6.625419494170049, "tokens_seen": 12648448 }, { "epoch": 0.0, "learning_rate": 0.00019246031746031748, "loss": 5.9043, "theoretical_loss": 6.619322467099223, "tokens_seen": 12713984 }, { "epoch": 0.0, "learning_rate": 0.00019345238095238096, "loss": 5.4898, "theoretical_loss": 6.613265535707211, "tokens_seen": 12779520 }, { "epoch": 0.0, "learning_rate": 0.00019444444444444446, "loss": 5.5153, "theoretical_loss": 6.607248232714213, "tokens_seen": 12845056 }, { "epoch": 0.0, "learning_rate": 0.00019543650793650793, "loss": 5.534, "theoretical_loss": 6.60127009863481, "tokens_seen": 12910592 }, { "epoch": 0.0, "learning_rate": 0.00019642857142857144, "loss": 5.7539, "theoretical_loss": 6.59533068160918, "tokens_seen": 12976128 }, { "epoch": 0.0, "learning_rate": 0.0001974206349206349, "loss": 5.633, "theoretical_loss": 6.589429537238785, "tokens_seen": 13041664 }, { "epoch": 0.0, "objective/train/docs_used": 19731, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.428110122680664, "objective/train/theoretical_loss": 6.583566228426414, "objective/train/tokens_used": 33567200, "theoretical_loss": 6.583566228426414, "tokens_seen": 13107200 }, { "epoch": 0.0, "learning_rate": 0.0001984126984126984, "loss": 5.3663, "theoretical_loss": 6.583566228426414, "tokens_seen": 13107200 }, { "epoch": 0.0, "learning_rate": 0.00019940476190476191, "loss": 5.7325, "theoretical_loss": 6.5777403252204305, "tokens_seen": 13172736 }, { "epoch": 0.0, "learning_rate": 0.00020039682539682542, "loss": 5.8639, "theoretical_loss": 6.571951404663098, "tokens_seen": 13238272 }, { "epoch": 0.0, "learning_rate": 0.0002013888888888889, "loss": 5.6695, "theoretical_loss": 6.566199050642863, "tokens_seen": 13303808 }, { "epoch": 0.0, "learning_rate": 0.0002023809523809524, "loss": 5.5443, "theoretical_loss": 6.560482853750463, "tokens_seen": 13369344 }, { "epoch": 0.0, "learning_rate": 0.00020337301587301587, "loss": 5.599, "theoretical_loss": 6.554802411138745, "tokens_seen": 13434880 }, { "epoch": 0.0, "learning_rate": 0.00020436507936507937, "loss": 5.3973, "theoretical_loss": 6.549157326386091, "tokens_seen": 13500416 }, { "epoch": 0.0, "learning_rate": 0.00020535714285714284, "loss": 5.6837, "theoretical_loss": 6.54354720936333, "tokens_seen": 13565952 }, { "epoch": 0.0, "learning_rate": 0.00020634920634920634, "loss": 5.4813, "theoretical_loss": 6.537971676104026, "tokens_seen": 13631488 }, { "epoch": 0.0, "learning_rate": 0.00020734126984126985, "loss": 5.6414, "theoretical_loss": 6.532430348678068, "tokens_seen": 13697024 }, { "epoch": 0.0, "learning_rate": 0.00020833333333333335, "loss": 5.5939, "theoretical_loss": 6.5269228550684195, "tokens_seen": 13762560 }, { "epoch": 0.0, "learning_rate": 0.00020932539682539685, "loss": 5.2664, "theoretical_loss": 6.521448829050978, "tokens_seen": 13828096 }, { "epoch": 0.0, "learning_rate": 0.00021031746031746032, "loss": 5.7298, "theoretical_loss": 6.516007910077416, "tokens_seen": 13893632 }, { "epoch": 0.0, "learning_rate": 0.00021130952380952382, "loss": 5.7504, "theoretical_loss": 6.51059974316095, "tokens_seen": 13959168 }, { "epoch": 0.0, "learning_rate": 0.0002123015873015873, "loss": 5.6754, "theoretical_loss": 6.50522397876491, "tokens_seen": 14024704 }, { "epoch": 0.0, "learning_rate": 0.0002132936507936508, "loss": 6.0764, "theoretical_loss": 6.499880272694068, "tokens_seen": 14090240 }, { "epoch": 0.0, "learning_rate": 0.00021428571428571427, "loss": 5.7123, "theoretical_loss": 6.494568285988618, "tokens_seen": 14155776 }, { "epoch": 0.0, "learning_rate": 0.0002152777777777778, "loss": 5.5138, "theoretical_loss": 6.489287684820745, "tokens_seen": 14221312 }, { "epoch": 0.0, "learning_rate": 0.00021626984126984128, "loss": 5.9094, "theoretical_loss": 6.484038140393699, "tokens_seen": 14286848 }, { "epoch": 0.0, "learning_rate": 0.00021726190476190478, "loss": 5.6079, "theoretical_loss": 6.4788193288433105, "tokens_seen": 14352384 }, { "epoch": 0.0, "learning_rate": 0.00021825396825396825, "loss": 5.5548, "theoretical_loss": 6.473630931141869, "tokens_seen": 14417920 }, { "epoch": 0.0, "learning_rate": 0.00021924603174603176, "loss": 5.5485, "theoretical_loss": 6.468472633004308, "tokens_seen": 14483456 }, { "epoch": 0.0, "learning_rate": 0.00022023809523809523, "loss": 5.6022, "theoretical_loss": 6.463344124796616, "tokens_seen": 14548992 }, { "epoch": 0.0, "learning_rate": 0.00022123015873015873, "loss": 5.715, "theoretical_loss": 6.45824510144643, "tokens_seen": 14614528 }, { "epoch": 0.0, "learning_rate": 0.0002222222222222222, "loss": 5.67, "theoretical_loss": 6.45317526235573, "tokens_seen": 14680064 }, { "epoch": 0.0, "objective/train/docs_used": 20428, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.015859127044678, "objective/train/theoretical_loss": 6.448134311315593, "objective/train/tokens_used": 35205600, "theoretical_loss": 6.448134311315593, "tokens_seen": 14745600 }, { "epoch": 0.0, "learning_rate": 0.00022321428571428573, "loss": 5.3803, "theoretical_loss": 6.448134311315593, "tokens_seen": 14745600 }, { "epoch": 0.0, "learning_rate": 0.0002242063492063492, "loss": 5.4398, "theoretical_loss": 6.443121956422939, "tokens_seen": 14811136 }, { "epoch": 0.0, "learning_rate": 0.0002251984126984127, "loss": 5.6742, "theoretical_loss": 6.438137909999214, "tokens_seen": 14876672 }, { "epoch": 0.0, "learning_rate": 0.00022619047619047618, "loss": 5.7337, "theoretical_loss": 6.433181888510964, "tokens_seen": 14942208 }, { "epoch": 0.0, "learning_rate": 0.00022718253968253969, "loss": 5.5962, "theoretical_loss": 6.428253612492239, "tokens_seen": 15007744 }, { "epoch": 0.0, "learning_rate": 0.00022817460317460316, "loss": 5.8581, "theoretical_loss": 6.4233528064687855, "tokens_seen": 15073280 }, { "epoch": 0.0, "learning_rate": 0.00022916666666666666, "loss": 5.7247, "theoretical_loss": 6.418479198883969, "tokens_seen": 15138816 }, { "epoch": 0.0, "learning_rate": 0.00023015873015873016, "loss": 5.6249, "theoretical_loss": 6.413632522026391, "tokens_seen": 15204352 }, { "epoch": 0.0, "learning_rate": 0.00023115079365079367, "loss": 5.6726, "theoretical_loss": 6.40881251195914, "tokens_seen": 15269888 }, { "epoch": 0.0, "learning_rate": 0.00023214285714285717, "loss": 5.4304, "theoretical_loss": 6.404018908450656, "tokens_seen": 15335424 }, { "epoch": 0.0, "learning_rate": 0.00023313492063492064, "loss": 5.4765, "theoretical_loss": 6.399251454907132, "tokens_seen": 15400960 }, { "epoch": 0.0, "learning_rate": 0.00023412698412698414, "loss": 5.865, "theoretical_loss": 6.394509898306452, "tokens_seen": 15466496 }, { "epoch": 0.0, "learning_rate": 0.00023511904761904762, "loss": 5.7674, "theoretical_loss": 6.389793989133574, "tokens_seen": 15532032 }, { "epoch": 0.0, "learning_rate": 0.00023611111111111112, "loss": 5.2223, "theoretical_loss": 6.385103481317387, "tokens_seen": 15597568 }, { "epoch": 0.0, "learning_rate": 0.0002371031746031746, "loss": 5.6511, "theoretical_loss": 6.380438132168923, "tokens_seen": 15663104 }, { "epoch": 0.0, "learning_rate": 0.0002380952380952381, "loss": 5.4302, "theoretical_loss": 6.375797702320966, "tokens_seen": 15728640 }, { "epoch": 0.0, "learning_rate": 0.0002390873015873016, "loss": 5.4526, "theoretical_loss": 6.371181955668966, "tokens_seen": 15794176 }, { "epoch": 0.0, "learning_rate": 0.0002400793650793651, "loss": 5.3307, "theoretical_loss": 6.366590659313248, "tokens_seen": 15859712 }, { "epoch": 0.0, "learning_rate": 0.00024107142857142857, "loss": 5.4796, "theoretical_loss": 6.36202358350248, "tokens_seen": 15925248 }, { "epoch": 0.0, "learning_rate": 0.00024206349206349207, "loss": 5.4457, "theoretical_loss": 6.357480501578371, "tokens_seen": 15990784 }, { "epoch": 0.0, "learning_rate": 0.00024305555555555555, "loss": 5.5368, "theoretical_loss": 6.352961189921553, "tokens_seen": 16056320 }, { "epoch": 0.0, "learning_rate": 0.00024404761904761905, "loss": 5.3346, "theoretical_loss": 6.348465427898629, "tokens_seen": 16121856 }, { "epoch": 0.0, "learning_rate": 0.00024503968253968255, "loss": 5.2158, "theoretical_loss": 6.343992997810366, "tokens_seen": 16187392 }, { "epoch": 0.0, "learning_rate": 0.000246031746031746, "loss": 5.5594, "theoretical_loss": 6.33954368484097, "tokens_seen": 16252928 }, { "epoch": 0.0, "learning_rate": 0.00024702380952380955, "loss": 5.5983, "theoretical_loss": 6.33511727700846, "tokens_seen": 16318464 }, { "epoch": 0.0, "objective/train/docs_used": 21641, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.571507930755615, "objective/train/theoretical_loss": 6.330713565116083, "objective/train/tokens_used": 36844000, "theoretical_loss": 6.330713565116083, "tokens_seen": 16384000 }, { "epoch": 0.0, "learning_rate": 0.000248015873015873, "loss": 5.6484, "theoretical_loss": 6.330713565116083, "tokens_seen": 16384000 }, { "epoch": 0.0, "learning_rate": 0.0002490079365079365, "loss": 5.6139, "theoretical_loss": 6.326332342704751, "tokens_seen": 16449536 }, { "epoch": 0.01, "learning_rate": 0.00025, "loss": 5.6404, "theoretical_loss": 6.32197340600647, "tokens_seen": 16515072 }, { "epoch": 0.01, "learning_rate": 0.0002509920634920635, "loss": 5.5749, "theoretical_loss": 6.3176365538987636, "tokens_seen": 16580608 }, { "epoch": 0.01, "learning_rate": 0.000251984126984127, "loss": 5.176, "theoretical_loss": 6.313321587860021, "tokens_seen": 16646144 }, { "epoch": 0.01, "learning_rate": 0.00025297619047619046, "loss": 5.4498, "theoretical_loss": 6.309028311925785, "tokens_seen": 16711680 }, { "epoch": 0.01, "learning_rate": 0.00025396825396825396, "loss": 5.059, "theoretical_loss": 6.304756532645939, "tokens_seen": 16777216 }, { "epoch": 0.01, "learning_rate": 0.00025496031746031746, "loss": 5.6538, "theoretical_loss": 6.300506059042775, "tokens_seen": 16842752 }, { "epoch": 0.01, "learning_rate": 0.00025595238095238096, "loss": 5.4578, "theoretical_loss": 6.296276702569918, "tokens_seen": 16908288 }, { "epoch": 0.01, "learning_rate": 0.0002569444444444444, "loss": 5.675, "theoretical_loss": 6.292068277072099, "tokens_seen": 16973824 }, { "epoch": 0.01, "learning_rate": 0.00025793650793650796, "loss": 5.6443, "theoretical_loss": 6.28788059874573, "tokens_seen": 17039360 }, { "epoch": 0.01, "learning_rate": 0.00025892857142857146, "loss": 4.9578, "theoretical_loss": 6.283713486100297, "tokens_seen": 17104896 }, { "epoch": 0.01, "learning_rate": 0.00025992063492063497, "loss": 5.6714, "theoretical_loss": 6.279566759920507, "tokens_seen": 17170432 }, { "epoch": 0.01, "learning_rate": 0.0002609126984126984, "loss": 5.4797, "theoretical_loss": 6.275440243229228, "tokens_seen": 17235968 }, { "epoch": 0.01, "learning_rate": 0.0002619047619047619, "loss": 5.2515, "theoretical_loss": 6.271333761251142, "tokens_seen": 17301504 }, { "epoch": 0.01, "learning_rate": 0.0002628968253968254, "loss": 5.4154, "theoretical_loss": 6.267247141377137, "tokens_seen": 17367040 }, { "epoch": 0.01, "learning_rate": 0.0002638888888888889, "loss": 5.5372, "theoretical_loss": 6.2631802131294085, "tokens_seen": 17432576 }, { "epoch": 0.01, "learning_rate": 0.00026488095238095237, "loss": 5.4489, "theoretical_loss": 6.259132808127246, "tokens_seen": 17498112 }, { "epoch": 0.01, "learning_rate": 0.00026587301587301587, "loss": 5.5914, "theoretical_loss": 6.255104760053497, "tokens_seen": 17563648 }, { "epoch": 0.01, "learning_rate": 0.00026686507936507937, "loss": 5.1736, "theoretical_loss": 6.251095904621689, "tokens_seen": 17629184 }, { "epoch": 0.01, "learning_rate": 0.00026785714285714287, "loss": 5.6947, "theoretical_loss": 6.247106079543801, "tokens_seen": 17694720 }, { "epoch": 0.01, "learning_rate": 0.0002688492063492063, "loss": 5.515, "theoretical_loss": 6.243135124498652, "tokens_seen": 17760256 }, { "epoch": 0.01, "learning_rate": 0.0002698412698412698, "loss": 5.531, "theoretical_loss": 6.239182881100916, "tokens_seen": 17825792 }, { "epoch": 0.01, "learning_rate": 0.0002708333333333333, "loss": 5.2666, "theoretical_loss": 6.235249192870732, "tokens_seen": 17891328 }, { "epoch": 0.01, "learning_rate": 0.0002718253968253968, "loss": 5.5261, "theoretical_loss": 6.231333905203899, "tokens_seen": 17956864 }, { "epoch": 0.01, "objective/train/docs_used": 22253, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.281318664550781, "objective/train/theoretical_loss": 6.227436865342643, "objective/train/tokens_used": 38482400, "theoretical_loss": 6.227436865342643, "tokens_seen": 18022400 }, { "epoch": 0.01, "learning_rate": 0.0002728174603174603, "loss": 5.2597, "theoretical_loss": 6.227436865342643, "tokens_seen": 18022400 }, { "epoch": 0.01, "learning_rate": 0.0002738095238095238, "loss": 5.6137, "theoretical_loss": 6.223557922346955, "tokens_seen": 18087936 }, { "epoch": 0.01, "learning_rate": 0.0002748015873015873, "loss": 5.2549, "theoretical_loss": 6.219696927066456, "tokens_seen": 18153472 }, { "epoch": 0.01, "learning_rate": 0.00027579365079365083, "loss": 5.3473, "theoretical_loss": 6.215853732112821, "tokens_seen": 18219008 }, { "epoch": 0.01, "learning_rate": 0.00027678571428571433, "loss": 5.2604, "theoretical_loss": 6.212028191832702, "tokens_seen": 18284544 }, { "epoch": 0.01, "learning_rate": 0.0002777777777777778, "loss": 5.3027, "theoretical_loss": 6.208220162281178, "tokens_seen": 18350080 }, { "epoch": 0.01, "learning_rate": 0.0002787698412698413, "loss": 5.2227, "theoretical_loss": 6.204429501195701, "tokens_seen": 18415616 }, { "epoch": 0.01, "learning_rate": 0.0002797619047619048, "loss": 5.4059, "theoretical_loss": 6.20065606797053, "tokens_seen": 18481152 }, { "epoch": 0.01, "learning_rate": 0.0002807539682539683, "loss": 5.5029, "theoretical_loss": 6.19689972363164, "tokens_seen": 18546688 }, { "epoch": 0.01, "learning_rate": 0.00028174603174603173, "loss": 5.3127, "theoretical_loss": 6.1931603308120975, "tokens_seen": 18612224 }, { "epoch": 0.01, "learning_rate": 0.00028273809523809523, "loss": 5.2913, "theoretical_loss": 6.189437753727901, "tokens_seen": 18677760 }, { "epoch": 0.01, "learning_rate": 0.00028373015873015873, "loss": 5.3535, "theoretical_loss": 6.185731858154261, "tokens_seen": 18743296 }, { "epoch": 0.01, "learning_rate": 0.00028472222222222223, "loss": 5.4967, "theoretical_loss": 6.182042511402313, "tokens_seen": 18808832 }, { "epoch": 0.01, "learning_rate": 0.0002857142857142857, "loss": 5.325, "theoretical_loss": 6.17836958229627, "tokens_seen": 18874368 }, { "epoch": 0.01, "learning_rate": 0.0002867063492063492, "loss": 5.3921, "theoretical_loss": 6.1747129411509825, "tokens_seen": 18939904 }, { "epoch": 0.01, "learning_rate": 0.0002876984126984127, "loss": 4.9391, "theoretical_loss": 6.171072459749913, "tokens_seen": 19005440 }, { "epoch": 0.01, "learning_rate": 0.0002886904761904762, "loss": 5.3633, "theoretical_loss": 6.1674480113235095, "tokens_seen": 19070976 }, { "epoch": 0.01, "learning_rate": 0.0002896825396825397, "loss": 5.2082, "theoretical_loss": 6.163839470527964, "tokens_seen": 19136512 }, { "epoch": 0.01, "learning_rate": 0.0002906746031746032, "loss": 5.3773, "theoretical_loss": 6.160246713424372, "tokens_seen": 19202048 }, { "epoch": 0.01, "learning_rate": 0.0002916666666666667, "loss": 5.7411, "theoretical_loss": 6.156669617458243, "tokens_seen": 19267584 }, { "epoch": 0.01, "learning_rate": 0.0002926587301587302, "loss": 4.9287, "theoretical_loss": 6.153108061439397, "tokens_seen": 19333120 }, { "epoch": 0.01, "learning_rate": 0.0002936507936507937, "loss": 5.493, "theoretical_loss": 6.149561925522211, "tokens_seen": 19398656 }, { "epoch": 0.01, "learning_rate": 0.00029464285714285714, "loss": 5.3644, "theoretical_loss": 6.146031091186222, "tokens_seen": 19464192 }, { "epoch": 0.01, "learning_rate": 0.00029563492063492064, "loss": 5.3852, "theoretical_loss": 6.142515441217064, "tokens_seen": 19529728 }, { "epoch": 0.01, "learning_rate": 0.00029662698412698414, "loss": 5.2977, "theoretical_loss": 6.1390148596877605, "tokens_seen": 19595264 }, { "epoch": 0.01, "objective/train/docs_used": 23612, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.15675163269043, "objective/train/theoretical_loss": 6.135529231940326, "objective/train/tokens_used": 40120800, "theoretical_loss": 6.135529231940326, "tokens_seen": 19660800 }, { "epoch": 0.01, "learning_rate": 0.00029761904761904765, "loss": 5.137, "theoretical_loss": 6.135529231940326, "tokens_seen": 19660800 }, { "epoch": 0.01, "learning_rate": 0.0002986111111111111, "loss": 5.1525, "theoretical_loss": 6.132058444567705, "tokens_seen": 19726336 }, { "epoch": 0.01, "learning_rate": 0.0002996031746031746, "loss": 5.0443, "theoretical_loss": 6.128602385396022, "tokens_seen": 19791872 }, { "epoch": 0.01, "learning_rate": 0.0003005952380952381, "loss": 5.4228, "theoretical_loss": 6.125160943467138, "tokens_seen": 19857408 }, { "epoch": 0.01, "learning_rate": 0.0003015873015873016, "loss": 5.194, "theoretical_loss": 6.121734009021521, "tokens_seen": 19922944 }, { "epoch": 0.01, "learning_rate": 0.00030257936507936505, "loss": 5.2838, "theoretical_loss": 6.118321473481398, "tokens_seen": 19988480 }, { "epoch": 0.01, "learning_rate": 0.00030357142857142855, "loss": 5.1757, "theoretical_loss": 6.114923229434213, "tokens_seen": 20054016 }, { "epoch": 0.01, "learning_rate": 0.00030456349206349205, "loss": 5.5533, "theoretical_loss": 6.111539170616359, "tokens_seen": 20119552 }, { "epoch": 0.01, "learning_rate": 0.0003055555555555556, "loss": 5.0377, "theoretical_loss": 6.108169191897195, "tokens_seen": 20185088 }, { "epoch": 0.01, "learning_rate": 0.00030654761904761905, "loss": 5.006, "theoretical_loss": 6.104813189263336, "tokens_seen": 20250624 }, { "epoch": 0.01, "learning_rate": 0.00030753968253968255, "loss": 5.2312, "theoretical_loss": 6.101471059803204, "tokens_seen": 20316160 }, { "epoch": 0.01, "learning_rate": 0.00030853174603174605, "loss": 5.2477, "theoretical_loss": 6.098142701691856, "tokens_seen": 20381696 }, { "epoch": 0.01, "learning_rate": 0.00030952380952380956, "loss": 5.1064, "theoretical_loss": 6.094828014176053, "tokens_seen": 20447232 }, { "epoch": 0.01, "learning_rate": 0.000310515873015873, "loss": 5.202, "theoretical_loss": 6.091526897559593, "tokens_seen": 20512768 }, { "epoch": 0.01, "learning_rate": 0.0003115079365079365, "loss": 5.0853, "theoretical_loss": 6.088239253188885, "tokens_seen": 20578304 }, { "epoch": 0.01, "learning_rate": 0.0003125, "loss": 5.3478, "theoretical_loss": 6.084964983438763, "tokens_seen": 20643840 }, { "epoch": 0.01, "learning_rate": 0.0003134920634920635, "loss": 5.2918, "theoretical_loss": 6.0817039916985465, "tokens_seen": 20709376 }, { "epoch": 0.01, "learning_rate": 0.000314484126984127, "loss": 5.0812, "theoretical_loss": 6.078456182358325, "tokens_seen": 20774912 }, { "epoch": 0.01, "learning_rate": 0.00031547619047619046, "loss": 5.1169, "theoretical_loss": 6.075221460795472, "tokens_seen": 20840448 }, { "epoch": 0.01, "learning_rate": 0.00031646825396825396, "loss": 5.279, "theoretical_loss": 6.071999733361386, "tokens_seen": 20905984 }, { "epoch": 0.01, "learning_rate": 0.00031746031746031746, "loss": 5.3881, "theoretical_loss": 6.068790907368448, "tokens_seen": 20971520 }, { "epoch": 0.01, "learning_rate": 0.00031845238095238096, "loss": 5.26, "theoretical_loss": 6.0655948910771915, "tokens_seen": 21037056 }, { "epoch": 0.01, "learning_rate": 0.0003194444444444444, "loss": 5.2783, "theoretical_loss": 6.062411593683687, "tokens_seen": 21102592 }, { "epoch": 0.01, "learning_rate": 0.00032043650793650796, "loss": 5.062, "theoretical_loss": 6.059240925307134, "tokens_seen": 21168128 }, { "epoch": 0.01, "learning_rate": 0.00032142857142857147, "loss": 5.2426, "theoretical_loss": 6.056082796977648, "tokens_seen": 21233664 }, { "epoch": 0.01, "objective/train/docs_used": 24330, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.736891746520996, "objective/train/theoretical_loss": 6.052937120624258, "objective/train/tokens_used": 41759200, "theoretical_loss": 6.052937120624258, "tokens_seen": 21299200 }, { "epoch": 0.01, "learning_rate": 0.00032242063492063497, "loss": 5.1217, "theoretical_loss": 6.052937120624258, "tokens_seen": 21299200 }, { "epoch": 0.01, "learning_rate": 0.0003234126984126984, "loss": 5.4246, "theoretical_loss": 6.049803809063083, "tokens_seen": 21364736 }, { "epoch": 0.01, "learning_rate": 0.0003244047619047619, "loss": 5.4049, "theoretical_loss": 6.0466827759857145, "tokens_seen": 21430272 }, { "epoch": 0.01, "learning_rate": 0.0003253968253968254, "loss": 5.1161, "theoretical_loss": 6.04357393594778, "tokens_seen": 21495808 }, { "epoch": 0.01, "learning_rate": 0.0003263888888888889, "loss": 5.2259, "theoretical_loss": 6.040477204357686, "tokens_seen": 21561344 }, { "epoch": 0.01, "learning_rate": 0.00032738095238095237, "loss": 5.4008, "theoretical_loss": 6.037392497465552, "tokens_seen": 21626880 }, { "epoch": 0.01, "learning_rate": 0.00032837301587301587, "loss": 5.0396, "theoretical_loss": 6.034319732352309, "tokens_seen": 21692416 }, { "epoch": 0.01, "learning_rate": 0.00032936507936507937, "loss": 5.1953, "theoretical_loss": 6.031258826918979, "tokens_seen": 21757952 }, { "epoch": 0.01, "learning_rate": 0.00033035714285714287, "loss": 5.3939, "theoretical_loss": 6.0282096998761245, "tokens_seen": 21823488 }, { "epoch": 0.01, "learning_rate": 0.0003313492063492063, "loss": 5.5128, "theoretical_loss": 6.025172270733464, "tokens_seen": 21889024 }, { "epoch": 0.01, "learning_rate": 0.0003323412698412698, "loss": 5.1908, "theoretical_loss": 6.0221464597896475, "tokens_seen": 21954560 }, { "epoch": 0.01, "learning_rate": 0.0003333333333333333, "loss": 4.7464, "theoretical_loss": 6.0191321881221995, "tokens_seen": 22020096 }, { "epoch": 0.01, "learning_rate": 0.0003343253968253968, "loss": 5.2092, "theoretical_loss": 6.016129377577614, "tokens_seen": 22085632 }, { "epoch": 0.01, "learning_rate": 0.0003353174603174603, "loss": 5.3966, "theoretical_loss": 6.01313795076161, "tokens_seen": 22151168 }, { "epoch": 0.01, "learning_rate": 0.0003363095238095238, "loss": 4.8061, "theoretical_loss": 6.010157831029533, "tokens_seen": 22216704 }, { "epoch": 0.01, "learning_rate": 0.00033730158730158733, "loss": 5.0669, "theoretical_loss": 6.007188942476907, "tokens_seen": 22282240 }, { "epoch": 0.01, "learning_rate": 0.00033829365079365083, "loss": 5.2141, "theoretical_loss": 6.0042312099301425, "tokens_seen": 22347776 }, { "epoch": 0.01, "learning_rate": 0.00033928571428571433, "loss": 5.1484, "theoretical_loss": 6.001284558937368, "tokens_seen": 22413312 }, { "epoch": 0.01, "learning_rate": 0.0003402777777777778, "loss": 5.2602, "theoretical_loss": 5.998348915759426, "tokens_seen": 22478848 }, { "epoch": 0.01, "learning_rate": 0.0003412698412698413, "loss": 5.1164, "theoretical_loss": 5.995424207360987, "tokens_seen": 22544384 }, { "epoch": 0.01, "learning_rate": 0.0003422619047619048, "loss": 5.0515, "theoretical_loss": 5.992510361401818, "tokens_seen": 22609920 }, { "epoch": 0.01, "learning_rate": 0.0003432539682539683, "loss": 5.218, "theoretical_loss": 5.989607306228168, "tokens_seen": 22675456 }, { "epoch": 0.01, "learning_rate": 0.00034424603174603173, "loss": 4.9102, "theoretical_loss": 5.986714970864292, "tokens_seen": 22740992 }, { "epoch": 0.01, "learning_rate": 0.00034523809523809523, "loss": 5.0805, "theoretical_loss": 5.983833285004112, "tokens_seen": 22806528 }, { "epoch": 0.01, "learning_rate": 0.00034623015873015873, "loss": 5.2809, "theoretical_loss": 5.980962179002983, "tokens_seen": 22872064 }, { "epoch": 0.01, "objective/train/docs_used": 24983, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.134525299072266, "objective/train/theoretical_loss": 5.978101583869607, "objective/train/tokens_used": 43397600, "theoretical_loss": 5.978101583869607, "tokens_seen": 22937600 }, { "epoch": 0.01, "learning_rate": 0.00034722222222222224, "loss": 5.1396, "theoretical_loss": 5.978101583869607, "tokens_seen": 22937600 }, { "epoch": 0.01, "learning_rate": 0.0003482142857142857, "loss": 5.0727, "theoretical_loss": 5.975251431258057, "tokens_seen": 23003136 }, { "epoch": 0.01, "learning_rate": 0.0003492063492063492, "loss": 5.1882, "theoretical_loss": 5.972411653459913, "tokens_seen": 23068672 }, { "epoch": 0.01, "learning_rate": 0.0003501984126984127, "loss": 4.8682, "theoretical_loss": 5.9695821833965335, "tokens_seen": 23134208 }, { "epoch": 0.01, "learning_rate": 0.0003511904761904762, "loss": 5.2343, "theoretical_loss": 5.966762954611432, "tokens_seen": 23199744 }, { "epoch": 0.01, "learning_rate": 0.0003521825396825397, "loss": 5.195, "theoretical_loss": 5.963953901262764, "tokens_seen": 23265280 }, { "epoch": 0.01, "learning_rate": 0.0003531746031746032, "loss": 5.2282, "theoretical_loss": 5.961154958115937, "tokens_seen": 23330816 }, { "epoch": 0.01, "learning_rate": 0.0003541666666666667, "loss": 4.7545, "theoretical_loss": 5.958366060536315, "tokens_seen": 23396352 }, { "epoch": 0.01, "learning_rate": 0.0003551587301587302, "loss": 5.1342, "theoretical_loss": 5.955587144482044, "tokens_seen": 23461888 }, { "epoch": 0.01, "learning_rate": 0.0003561507936507937, "loss": 5.0144, "theoretical_loss": 5.952818146496978, "tokens_seen": 23527424 }, { "epoch": 0.01, "learning_rate": 0.00035714285714285714, "loss": 5.0659, "theoretical_loss": 5.950059003703704, "tokens_seen": 23592960 }, { "epoch": 0.01, "learning_rate": 0.00035813492063492064, "loss": 5.1206, "theoretical_loss": 5.94730965379668, "tokens_seen": 23658496 }, { "epoch": 0.01, "learning_rate": 0.00035912698412698415, "loss": 4.9633, "theoretical_loss": 5.944570035035458, "tokens_seen": 23724032 }, { "epoch": 0.01, "learning_rate": 0.00036011904761904765, "loss": 5.1591, "theoretical_loss": 5.941840086238027, "tokens_seen": 23789568 }, { "epoch": 0.01, "learning_rate": 0.0003611111111111111, "loss": 5.3314, "theoretical_loss": 5.939119746774228, "tokens_seen": 23855104 }, { "epoch": 0.01, "learning_rate": 0.0003621031746031746, "loss": 5.2119, "theoretical_loss": 5.936408956559284, "tokens_seen": 23920640 }, { "epoch": 0.01, "learning_rate": 0.0003630952380952381, "loss": 5.3978, "theoretical_loss": 5.933707656047414, "tokens_seen": 23986176 }, { "epoch": 0.01, "learning_rate": 0.0003640873015873016, "loss": 4.9459, "theoretical_loss": 5.93101578622554, "tokens_seen": 24051712 }, { "epoch": 0.01, "learning_rate": 0.00036507936507936505, "loss": 5.031, "theoretical_loss": 5.928333288607086, "tokens_seen": 24117248 }, { "epoch": 0.01, "learning_rate": 0.00036607142857142855, "loss": 5.1772, "theoretical_loss": 5.925660105225867, "tokens_seen": 24182784 }, { "epoch": 0.01, "learning_rate": 0.00036706349206349205, "loss": 4.8824, "theoretical_loss": 5.92299617863006, "tokens_seen": 24248320 }, { "epoch": 0.01, "learning_rate": 0.0003680555555555556, "loss": 5.0966, "theoretical_loss": 5.920341451876267, "tokens_seen": 24313856 }, { "epoch": 0.01, "learning_rate": 0.00036904761904761905, "loss": 5.1533, "theoretical_loss": 5.9176958685236585, "tokens_seen": 24379392 }, { "epoch": 0.01, "learning_rate": 0.00037003968253968255, "loss": 5.2939, "theoretical_loss": 5.9150593726282015, "tokens_seen": 24444928 }, { "epoch": 0.01, "learning_rate": 0.00037103174603174606, "loss": 5.0167, "theoretical_loss": 5.912431908736972, "tokens_seen": 24510464 }, { "epoch": 0.01, "objective/train/docs_used": 26439, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.887441635131836, "objective/train/theoretical_loss": 5.909813421882534, "objective/train/tokens_used": 45036000, "theoretical_loss": 5.909813421882534, "tokens_seen": 24576000 }, { "epoch": 0.01, "learning_rate": 0.00037202380952380956, "loss": 5.2049, "theoretical_loss": 5.909813421882534, "tokens_seen": 24576000 }, { "epoch": 0.01, "learning_rate": 0.000373015873015873, "loss": 5.2346, "theoretical_loss": 5.907203857577422, "tokens_seen": 24641536 }, { "epoch": 0.01, "learning_rate": 0.0003740079365079365, "loss": 5.2112, "theoretical_loss": 5.9046031618086765, "tokens_seen": 24707072 }, { "epoch": 0.01, "learning_rate": 0.000375, "loss": 5.092, "theoretical_loss": 5.902011281032472, "tokens_seen": 24772608 }, { "epoch": 0.01, "learning_rate": 0.0003759920634920635, "loss": 5.2228, "theoretical_loss": 5.899428162168808, "tokens_seen": 24838144 }, { "epoch": 0.01, "learning_rate": 0.000376984126984127, "loss": 4.9695, "theoretical_loss": 5.896853752596286, "tokens_seen": 24903680 }, { "epoch": 0.01, "learning_rate": 0.00037797619047619046, "loss": 5.1379, "theoretical_loss": 5.894288000146949, "tokens_seen": 24969216 }, { "epoch": 0.01, "learning_rate": 0.00037896825396825396, "loss": 5.0639, "theoretical_loss": 5.891730853101199, "tokens_seen": 25034752 }, { "epoch": 0.01, "learning_rate": 0.00037996031746031746, "loss": 5.0119, "theoretical_loss": 5.88918226018278, "tokens_seen": 25100288 }, { "epoch": 0.01, "learning_rate": 0.00038095238095238096, "loss": 5.0625, "theoretical_loss": 5.8866421705538325, "tokens_seen": 25165824 }, { "epoch": 0.01, "learning_rate": 0.0003819444444444444, "loss": 5.2315, "theoretical_loss": 5.8841105338100155, "tokens_seen": 25231360 }, { "epoch": 0.01, "learning_rate": 0.00038293650793650797, "loss": 4.8079, "theoretical_loss": 5.881587299975694, "tokens_seen": 25296896 }, { "epoch": 0.01, "learning_rate": 0.00038392857142857147, "loss": 5.3085, "theoretical_loss": 5.8790724194991935, "tokens_seen": 25362432 }, { "epoch": 0.01, "learning_rate": 0.00038492063492063497, "loss": 5.1826, "theoretical_loss": 5.876565843248124, "tokens_seen": 25427968 }, { "epoch": 0.01, "learning_rate": 0.0003859126984126984, "loss": 4.8865, "theoretical_loss": 5.8740675225047525, "tokens_seen": 25493504 }, { "epoch": 0.01, "learning_rate": 0.0003869047619047619, "loss": 4.8372, "theoretical_loss": 5.871577408961457, "tokens_seen": 25559040 }, { "epoch": 0.01, "learning_rate": 0.0003878968253968254, "loss": 5.1074, "theoretical_loss": 5.869095454716231, "tokens_seen": 25624576 }, { "epoch": 0.01, "learning_rate": 0.0003888888888888889, "loss": 5.0627, "theoretical_loss": 5.866621612268246, "tokens_seen": 25690112 }, { "epoch": 0.01, "learning_rate": 0.00038988095238095237, "loss": 5.1708, "theoretical_loss": 5.864155834513486, "tokens_seen": 25755648 }, { "epoch": 0.01, "learning_rate": 0.00039087301587301587, "loss": 4.9657, "theoretical_loss": 5.8616980747404295, "tokens_seen": 25821184 }, { "epoch": 0.01, "learning_rate": 0.00039186507936507937, "loss": 5.1406, "theoretical_loss": 5.859248286625787, "tokens_seen": 25886720 }, { "epoch": 0.01, "learning_rate": 0.0003928571428571429, "loss": 5.253, "theoretical_loss": 5.856806424230314, "tokens_seen": 25952256 }, { "epoch": 0.01, "learning_rate": 0.0003938492063492063, "loss": 4.9518, "theoretical_loss": 5.854372441994654, "tokens_seen": 26017792 }, { "epoch": 0.01, "learning_rate": 0.0003948412698412698, "loss": 4.9569, "theoretical_loss": 5.851946294735258, "tokens_seen": 26083328 }, { "epoch": 0.01, "learning_rate": 0.0003958333333333333, "loss": 4.9376, "theoretical_loss": 5.849527937640345, "tokens_seen": 26148864 }, { "epoch": 0.01, "objective/train/docs_used": 27060, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.355391025543213, "objective/train/theoretical_loss": 5.8471173262659235, "objective/train/tokens_used": 46674400, "theoretical_loss": 5.8471173262659235, "tokens_seen": 26214400 }, { "epoch": 0.01, "learning_rate": 0.0003968253968253968, "loss": 4.8582, "theoretical_loss": 5.8471173262659235, "tokens_seen": 26214400 }, { "epoch": 0.01, "learning_rate": 0.0003978174603174603, "loss": 4.9394, "theoretical_loss": 5.84471441653186, "tokens_seen": 26279936 }, { "epoch": 0.01, "learning_rate": 0.00039880952380952383, "loss": 5.2367, "theoretical_loss": 5.842319164718004, "tokens_seen": 26345472 }, { "epoch": 0.01, "learning_rate": 0.00039980158730158733, "loss": 5.1756, "theoretical_loss": 5.83993152746036, "tokens_seen": 26411008 }, { "epoch": 0.01, "learning_rate": 0.00040079365079365083, "loss": 5.1252, "theoretical_loss": 5.83755146174731, "tokens_seen": 26476544 }, { "epoch": 0.01, "learning_rate": 0.00040178571428571433, "loss": 5.0547, "theoretical_loss": 5.835178924915889, "tokens_seen": 26542080 }, { "epoch": 0.01, "learning_rate": 0.0004027777777777778, "loss": 5.0688, "theoretical_loss": 5.832813874648102, "tokens_seen": 26607616 }, { "epoch": 0.01, "learning_rate": 0.0004037698412698413, "loss": 5.0608, "theoretical_loss": 5.8304562689673, "tokens_seen": 26673152 }, { "epoch": 0.01, "learning_rate": 0.0004047619047619048, "loss": 5.0401, "theoretical_loss": 5.828106066234588, "tokens_seen": 26738688 }, { "epoch": 0.01, "learning_rate": 0.0004057539682539683, "loss": 5.0167, "theoretical_loss": 5.825763225145295, "tokens_seen": 26804224 }, { "epoch": 0.01, "learning_rate": 0.00040674603174603173, "loss": 5.1791, "theoretical_loss": 5.823427704725473, "tokens_seen": 26869760 }, { "epoch": 0.01, "learning_rate": 0.00040773809523809523, "loss": 4.8063, "theoretical_loss": 5.82109946432846, "tokens_seen": 26935296 }, { "epoch": 0.01, "learning_rate": 0.00040873015873015874, "loss": 4.9278, "theoretical_loss": 5.818778463631473, "tokens_seen": 27000832 }, { "epoch": 0.01, "learning_rate": 0.00040972222222222224, "loss": 4.7786, "theoretical_loss": 5.816464662632243, "tokens_seen": 27066368 }, { "epoch": 0.01, "learning_rate": 0.0004107142857142857, "loss": 5.1196, "theoretical_loss": 5.8141580216457065, "tokens_seen": 27131904 }, { "epoch": 0.01, "learning_rate": 0.0004117063492063492, "loss": 4.6589, "theoretical_loss": 5.811858501300729, "tokens_seen": 27197440 }, { "epoch": 0.01, "learning_rate": 0.0004126984126984127, "loss": 4.8227, "theoretical_loss": 5.809566062536868, "tokens_seen": 27262976 }, { "epoch": 0.01, "learning_rate": 0.0004136904761904762, "loss": 5.0245, "theoretical_loss": 5.807280666601191, "tokens_seen": 27328512 }, { "epoch": 0.01, "learning_rate": 0.0004146825396825397, "loss": 4.9783, "theoretical_loss": 5.805002275045111, "tokens_seen": 27394048 }, { "epoch": 0.01, "learning_rate": 0.0004156746031746032, "loss": 5.0457, "theoretical_loss": 5.8027308497212875, "tokens_seen": 27459584 }, { "epoch": 0.01, "learning_rate": 0.0004166666666666667, "loss": 5.002, "theoretical_loss": 5.800466352780546, "tokens_seen": 27525120 }, { "epoch": 0.01, "learning_rate": 0.0004176587301587302, "loss": 4.7979, "theoretical_loss": 5.798208746668847, "tokens_seen": 27590656 }, { "epoch": 0.01, "learning_rate": 0.0004186507936507937, "loss": 5.0751, "theoretical_loss": 5.795957994124291, "tokens_seen": 27656192 }, { "epoch": 0.01, "learning_rate": 0.00041964285714285714, "loss": 4.5446, "theoretical_loss": 5.7937140581741575, "tokens_seen": 27721728 }, { "epoch": 0.01, "learning_rate": 0.00042063492063492065, "loss": 4.9859, "theoretical_loss": 5.791476902131985, "tokens_seen": 27787264 }, { "epoch": 0.01, "objective/train/docs_used": 28453, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.114999294281006, "objective/train/theoretical_loss": 5.789246489594688, "objective/train/tokens_used": 48312800, "theoretical_loss": 5.789246489594688, "tokens_seen": 27852800 }, { "epoch": 0.01, "learning_rate": 0.00042162698412698415, "loss": 4.7322, "theoretical_loss": 5.789246489594688, "tokens_seen": 27852800 }, { "epoch": 0.01, "learning_rate": 0.00042261904761904765, "loss": 4.9798, "theoretical_loss": 5.787022784439701, "tokens_seen": 27918336 }, { "epoch": 0.01, "learning_rate": 0.0004236111111111111, "loss": 5.1612, "theoretical_loss": 5.784805750822171, "tokens_seen": 27983872 }, { "epoch": 0.01, "learning_rate": 0.0004246031746031746, "loss": 5.0818, "theoretical_loss": 5.782595353172176, "tokens_seen": 28049408 }, { "epoch": 0.01, "learning_rate": 0.0004255952380952381, "loss": 4.5939, "theoretical_loss": 5.780391556191977, "tokens_seen": 28114944 }, { "epoch": 0.01, "learning_rate": 0.0004265873015873016, "loss": 4.9998, "theoretical_loss": 5.778194324853311, "tokens_seen": 28180480 }, { "epoch": 0.01, "learning_rate": 0.00042757936507936505, "loss": 5.1566, "theoretical_loss": 5.776003624394711, "tokens_seen": 28246016 }, { "epoch": 0.01, "learning_rate": 0.00042857142857142855, "loss": 5.2049, "theoretical_loss": 5.773819420318858, "tokens_seen": 28311552 }, { "epoch": 0.01, "learning_rate": 0.00042956349206349205, "loss": 4.7744, "theoretical_loss": 5.771641678389971, "tokens_seen": 28377088 }, { "epoch": 0.01, "learning_rate": 0.0004305555555555556, "loss": 4.8862, "theoretical_loss": 5.769470364631225, "tokens_seen": 28442624 }, { "epoch": 0.01, "learning_rate": 0.00043154761904761905, "loss": 4.7604, "theoretical_loss": 5.767305445322201, "tokens_seen": 28508160 }, { "epoch": 0.01, "learning_rate": 0.00043253968253968256, "loss": 4.9808, "theoretical_loss": 5.765146886996363, "tokens_seen": 28573696 }, { "epoch": 0.01, "learning_rate": 0.00043353174603174606, "loss": 4.7763, "theoretical_loss": 5.762994656438579, "tokens_seen": 28639232 }, { "epoch": 0.01, "learning_rate": 0.00043452380952380956, "loss": 5.0911, "theoretical_loss": 5.760848720682651, "tokens_seen": 28704768 }, { "epoch": 0.01, "learning_rate": 0.000435515873015873, "loss": 5.0289, "theoretical_loss": 5.758709047008894, "tokens_seen": 28770304 }, { "epoch": 0.01, "learning_rate": 0.0004365079365079365, "loss": 5.0362, "theoretical_loss": 5.756575602941732, "tokens_seen": 28835840 }, { "epoch": 0.01, "learning_rate": 0.0004375, "loss": 5.1589, "theoretical_loss": 5.75444835624733, "tokens_seen": 28901376 }, { "epoch": 0.01, "learning_rate": 0.0004384920634920635, "loss": 5.1524, "theoretical_loss": 5.752327274931249, "tokens_seen": 28966912 }, { "epoch": 0.01, "learning_rate": 0.000439484126984127, "loss": 4.6286, "theoretical_loss": 5.750212327236129, "tokens_seen": 29032448 }, { "epoch": 0.01, "learning_rate": 0.00044047619047619046, "loss": 4.9918, "theoretical_loss": 5.7481034816394105, "tokens_seen": 29097984 }, { "epoch": 0.01, "learning_rate": 0.00044146825396825396, "loss": 5.0182, "theoretical_loss": 5.7460007068510635, "tokens_seen": 29163520 }, { "epoch": 0.01, "learning_rate": 0.00044246031746031746, "loss": 5.1442, "theoretical_loss": 5.74390397181136, "tokens_seen": 29229056 }, { "epoch": 0.01, "learning_rate": 0.00044345238095238096, "loss": 4.9284, "theoretical_loss": 5.741813245688668, "tokens_seen": 29294592 }, { "epoch": 0.01, "learning_rate": 0.0004444444444444444, "loss": 4.7274, "theoretical_loss": 5.739728497877267, "tokens_seen": 29360128 }, { "epoch": 0.01, "learning_rate": 0.00044543650793650797, "loss": 4.9829, "theoretical_loss": 5.737649697995197, "tokens_seen": 29425664 }, { "epoch": 0.01, "objective/train/docs_used": 29115, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.875042915344238, "objective/train/theoretical_loss": 5.7355768158821245, "objective/train/tokens_used": 49951200, "theoretical_loss": 5.7355768158821245, "tokens_seen": 29491200 }, { "epoch": 0.01, "learning_rate": 0.00044642857142857147, "loss": 5.0802, "theoretical_loss": 5.7355768158821245, "tokens_seen": 29491200 }, { "epoch": 0.01, "learning_rate": 0.00044742063492063497, "loss": 4.8054, "theoretical_loss": 5.73350982159724, "tokens_seen": 29556736 }, { "epoch": 0.01, "learning_rate": 0.0004484126984126984, "loss": 4.9775, "theoretical_loss": 5.731448685417178, "tokens_seen": 29622272 }, { "epoch": 0.01, "learning_rate": 0.0004494047619047619, "loss": 4.9123, "theoretical_loss": 5.729393377833956, "tokens_seen": 29687808 }, { "epoch": 0.01, "learning_rate": 0.0004503968253968254, "loss": 4.5612, "theoretical_loss": 5.7273438695529535, "tokens_seen": 29753344 }, { "epoch": 0.01, "learning_rate": 0.0004513888888888889, "loss": 5.0027, "theoretical_loss": 5.725300131490888, "tokens_seen": 29818880 }, { "epoch": 0.01, "learning_rate": 0.00045238095238095237, "loss": 4.9368, "theoretical_loss": 5.7232621347738455, "tokens_seen": 29884416 }, { "epoch": 0.01, "learning_rate": 0.00045337301587301587, "loss": 5.0227, "theoretical_loss": 5.721229850735305, "tokens_seen": 29949952 }, { "epoch": 0.01, "learning_rate": 0.00045436507936507937, "loss": 5.0473, "theoretical_loss": 5.719203250914208, "tokens_seen": 30015488 }, { "epoch": 0.01, "learning_rate": 0.0004553571428571429, "loss": 5.0973, "theoretical_loss": 5.717182307053037, "tokens_seen": 30081024 }, { "epoch": 0.01, "learning_rate": 0.0004563492063492063, "loss": 4.757, "theoretical_loss": 5.715166991095922, "tokens_seen": 30146560 }, { "epoch": 0.01, "learning_rate": 0.0004573412698412698, "loss": 4.8415, "theoretical_loss": 5.713157275186761, "tokens_seen": 30212096 }, { "epoch": 0.01, "learning_rate": 0.0004583333333333333, "loss": 4.9583, "theoretical_loss": 5.71115313166738, "tokens_seen": 30277632 }, { "epoch": 0.01, "learning_rate": 0.0004593253968253968, "loss": 5.1126, "theoretical_loss": 5.709154533075688, "tokens_seen": 30343168 }, { "epoch": 0.01, "learning_rate": 0.00046031746031746033, "loss": 4.9564, "theoretical_loss": 5.707161452143879, "tokens_seen": 30408704 }, { "epoch": 0.01, "learning_rate": 0.00046130952380952383, "loss": 5.0017, "theoretical_loss": 5.7051738617966326, "tokens_seen": 30474240 }, { "epoch": 0.01, "learning_rate": 0.00046230158730158733, "loss": 5.1598, "theoretical_loss": 5.7031917351493515, "tokens_seen": 30539776 }, { "epoch": 0.01, "learning_rate": 0.00046329365079365083, "loss": 4.7815, "theoretical_loss": 5.701215045506411, "tokens_seen": 30605312 }, { "epoch": 0.01, "learning_rate": 0.00046428571428571433, "loss": 5.0404, "theoretical_loss": 5.699243766359421, "tokens_seen": 30670848 }, { "epoch": 0.01, "learning_rate": 0.0004652777777777778, "loss": 4.958, "theoretical_loss": 5.697277871385534, "tokens_seen": 30736384 }, { "epoch": 0.01, "learning_rate": 0.0004662698412698413, "loss": 4.8348, "theoretical_loss": 5.695317334445736, "tokens_seen": 30801920 }, { "epoch": 0.01, "learning_rate": 0.0004672619047619048, "loss": 5.1101, "theoretical_loss": 5.693362129583184, "tokens_seen": 30867456 }, { "epoch": 0.01, "learning_rate": 0.0004682539682539683, "loss": 5.1583, "theoretical_loss": 5.691412231021549, "tokens_seen": 30932992 }, { "epoch": 0.01, "learning_rate": 0.00046924603174603173, "loss": 5.0497, "theoretical_loss": 5.689467613163388, "tokens_seen": 30998528 }, { "epoch": 0.01, "learning_rate": 0.00047023809523809523, "loss": 5.0994, "theoretical_loss": 5.687528250588518, "tokens_seen": 31064064 }, { "epoch": 0.01, "objective/train/docs_used": 30152, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.741349697113037, "objective/train/theoretical_loss": 5.6855941180524265, "objective/train/tokens_used": 51589600, "theoretical_loss": 5.6855941180524265, "tokens_seen": 31129600 }, { "epoch": 0.01, "learning_rate": 0.00047123015873015874, "loss": 4.9423, "theoretical_loss": 5.6855941180524265, "tokens_seen": 31129600 }, { "epoch": 0.01, "learning_rate": 0.00047222222222222224, "loss": 4.9094, "theoretical_loss": 5.683665190484683, "tokens_seen": 31195136 }, { "epoch": 0.01, "learning_rate": 0.0004732142857142857, "loss": 5.0356, "theoretical_loss": 5.681741442987381, "tokens_seen": 31260672 }, { "epoch": 0.01, "learning_rate": 0.0004742063492063492, "loss": 4.8249, "theoretical_loss": 5.679822850833591, "tokens_seen": 31326208 }, { "epoch": 0.01, "learning_rate": 0.0004751984126984127, "loss": 4.9261, "theoretical_loss": 5.677909389465831, "tokens_seen": 31391744 }, { "epoch": 0.01, "learning_rate": 0.0004761904761904762, "loss": 4.5377, "theoretical_loss": 5.676001034494554, "tokens_seen": 31457280 }, { "epoch": 0.01, "learning_rate": 0.0004771825396825397, "loss": 4.464, "theoretical_loss": 5.674097761696653, "tokens_seen": 31522816 }, { "epoch": 0.01, "learning_rate": 0.0004781746031746032, "loss": 4.6988, "theoretical_loss": 5.672199547013983, "tokens_seen": 31588352 }, { "epoch": 0.01, "learning_rate": 0.0004791666666666667, "loss": 4.6733, "theoretical_loss": 5.670306366551898, "tokens_seen": 31653888 }, { "epoch": 0.01, "learning_rate": 0.0004801587301587302, "loss": 4.7771, "theoretical_loss": 5.6684181965778, "tokens_seen": 31719424 }, { "epoch": 0.01, "learning_rate": 0.0004811507936507937, "loss": 5.0062, "theoretical_loss": 5.666535013519715, "tokens_seen": 31784960 }, { "epoch": 0.01, "learning_rate": 0.00048214285714285715, "loss": 4.7684, "theoretical_loss": 5.6646567939648715, "tokens_seen": 31850496 }, { "epoch": 0.01, "learning_rate": 0.00048313492063492065, "loss": 4.8464, "theoretical_loss": 5.6627835146583045, "tokens_seen": 31916032 }, { "epoch": 0.01, "learning_rate": 0.00048412698412698415, "loss": 4.689, "theoretical_loss": 5.660915152501465, "tokens_seen": 31981568 }, { "epoch": 0.01, "learning_rate": 0.00048511904761904765, "loss": 4.5762, "theoretical_loss": 5.659051684550857, "tokens_seen": 32047104 }, { "epoch": 0.01, "learning_rate": 0.0004861111111111111, "loss": 4.8833, "theoretical_loss": 5.657193088016677, "tokens_seen": 32112640 }, { "epoch": 0.01, "learning_rate": 0.0004871031746031746, "loss": 4.6286, "theoretical_loss": 5.655339340261474, "tokens_seen": 32178176 }, { "epoch": 0.01, "learning_rate": 0.0004880952380952381, "loss": 4.791, "theoretical_loss": 5.653490418798825, "tokens_seen": 32243712 }, { "epoch": 0.01, "learning_rate": 0.0004890873015873016, "loss": 5.0112, "theoretical_loss": 5.651646301292022, "tokens_seen": 32309248 }, { "epoch": 0.01, "learning_rate": 0.0004900793650793651, "loss": 4.7112, "theoretical_loss": 5.649806965552774, "tokens_seen": 32374784 }, { "epoch": 0.01, "learning_rate": 0.0004910714285714286, "loss": 4.7717, "theoretical_loss": 5.6479723895399205, "tokens_seen": 32440320 }, { "epoch": 0.01, "learning_rate": 0.000492063492063492, "loss": 4.7971, "theoretical_loss": 5.6461425513581665, "tokens_seen": 32505856 }, { "epoch": 0.01, "learning_rate": 0.0004930555555555556, "loss": 4.9546, "theoretical_loss": 5.6443174292568195, "tokens_seen": 32571392 }, { "epoch": 0.01, "learning_rate": 0.0004940476190476191, "loss": 4.6853, "theoretical_loss": 5.6424970016285485, "tokens_seen": 32636928 }, { "epoch": 0.01, "learning_rate": 0.0004950396825396826, "loss": 4.8721, "theoretical_loss": 5.640681247008156, "tokens_seen": 32702464 }, { "epoch": 0.01, "objective/train/docs_used": 30758, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.8071393966674805, "objective/train/theoretical_loss": 5.638870144071353, "objective/train/tokens_used": 53228000, "theoretical_loss": 5.638870144071353, "tokens_seen": 32768000 }, { "epoch": 0.01, "learning_rate": 0.000496031746031746, "loss": 4.8628, "theoretical_loss": 5.638870144071353, "tokens_seen": 32768000 }, { "epoch": 0.01, "learning_rate": 0.0004970238095238095, "loss": 4.7892, "theoretical_loss": 5.637063671633564, "tokens_seen": 32833536 }, { "epoch": 0.01, "learning_rate": 0.000498015873015873, "loss": 4.8607, "theoretical_loss": 5.635261808648728, "tokens_seen": 32899072 }, { "epoch": 0.01, "learning_rate": 0.0004990079365079365, "loss": 5.0888, "theoretical_loss": 5.6334645342081195, "tokens_seen": 32964608 }, { "epoch": 0.01, "learning_rate": 0.0005, "loss": 4.9737, "theoretical_loss": 5.631671827539186, "tokens_seen": 33030144 }, { "epoch": 0.01, "learning_rate": 0.0004999899699097292, "loss": 5.0126, "theoretical_loss": 5.629883668004389, "tokens_seen": 33095680 }, { "epoch": 0.01, "learning_rate": 0.0004999799398194584, "loss": 4.9297, "theoretical_loss": 5.628100035100061, "tokens_seen": 33161216 }, { "epoch": 0.01, "learning_rate": 0.0004999699097291876, "loss": 5.0938, "theoretical_loss": 5.626320908455279, "tokens_seen": 33226752 }, { "epoch": 0.01, "learning_rate": 0.0004999598796389167, "loss": 4.758, "theoretical_loss": 5.6245462678307385, "tokens_seen": 33292288 }, { "epoch": 0.01, "learning_rate": 0.000499949849548646, "loss": 4.6107, "theoretical_loss": 5.622776093117652, "tokens_seen": 33357824 }, { "epoch": 0.01, "learning_rate": 0.0004999398194583751, "loss": 4.9094, "theoretical_loss": 5.621010364336651, "tokens_seen": 33423360 }, { "epoch": 0.01, "learning_rate": 0.0004999297893681044, "loss": 4.7434, "theoretical_loss": 5.619249061636698, "tokens_seen": 33488896 }, { "epoch": 0.01, "learning_rate": 0.0004999197592778335, "loss": 4.7376, "theoretical_loss": 5.61749216529402, "tokens_seen": 33554432 }, { "epoch": 0.01, "learning_rate": 0.0004999097291875627, "loss": 4.8733, "theoretical_loss": 5.615739655711037, "tokens_seen": 33619968 }, { "epoch": 0.01, "learning_rate": 0.0004998996990972919, "loss": 4.8815, "theoretical_loss": 5.61399151341532, "tokens_seen": 33685504 }, { "epoch": 0.01, "learning_rate": 0.0004998896690070211, "loss": 4.8804, "theoretical_loss": 5.6122477190585425, "tokens_seen": 33751040 }, { "epoch": 0.01, "learning_rate": 0.0004998796389167503, "loss": 4.6509, "theoretical_loss": 5.610508253415453, "tokens_seen": 33816576 }, { "epoch": 0.01, "learning_rate": 0.0004998696088264795, "loss": 5.0255, "theoretical_loss": 5.6087730973828585, "tokens_seen": 33882112 }, { "epoch": 0.01, "learning_rate": 0.0004998595787362087, "loss": 4.7137, "theoretical_loss": 5.6070422319786095, "tokens_seen": 33947648 }, { "epoch": 0.01, "learning_rate": 0.0004998495486459378, "loss": 4.8675, "theoretical_loss": 5.605315638340606, "tokens_seen": 34013184 }, { "epoch": 0.01, "learning_rate": 0.000499839518555667, "loss": 4.8488, "theoretical_loss": 5.603593297725807, "tokens_seen": 34078720 }, { "epoch": 0.01, "learning_rate": 0.0004998294884653962, "loss": 5.0848, "theoretical_loss": 5.601875191509249, "tokens_seen": 34144256 }, { "epoch": 0.01, "learning_rate": 0.0004998194583751254, "loss": 4.4149, "theoretical_loss": 5.600161301183084, "tokens_seen": 34209792 }, { "epoch": 0.01, "learning_rate": 0.0004998094282848546, "loss": 5.0088, "theoretical_loss": 5.598451608355614, "tokens_seen": 34275328 }, { "epoch": 0.01, "learning_rate": 0.0004997993981945837, "loss": 4.9221, "theoretical_loss": 5.596746094750342, "tokens_seen": 34340864 }, { "epoch": 0.01, "objective/train/docs_used": 31355, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 5.054786205291748, "objective/train/theoretical_loss": 5.595044742205037, "objective/train/tokens_used": 54866400, "theoretical_loss": 5.595044742205037, "tokens_seen": 34406400 }, { "epoch": 0.01, "learning_rate": 0.0004997893681043129, "loss": 5.0983, "theoretical_loss": 5.595044742205037, "tokens_seen": 34406400 }, { "epoch": 0.01, "learning_rate": 0.0004997793380140421, "loss": 4.5808, "theoretical_loss": 5.5933475326707995, "tokens_seen": 34471936 }, { "epoch": 0.01, "learning_rate": 0.0004997693079237714, "loss": 4.5657, "theoretical_loss": 5.591654448211143, "tokens_seen": 34537472 }, { "epoch": 0.01, "learning_rate": 0.0004997592778335005, "loss": 4.8711, "theoretical_loss": 5.589965471001077, "tokens_seen": 34603008 }, { "epoch": 0.01, "learning_rate": 0.0004997492477432298, "loss": 4.8155, "theoretical_loss": 5.5882805833262115, "tokens_seen": 34668544 }, { "epoch": 0.01, "learning_rate": 0.0004997392176529588, "loss": 4.9002, "theoretical_loss": 5.586599767581859, "tokens_seen": 34734080 }, { "epoch": 0.01, "learning_rate": 0.0004997291875626881, "loss": 4.9815, "theoretical_loss": 5.584923006272151, "tokens_seen": 34799616 }, { "epoch": 0.01, "learning_rate": 0.0004997191574724173, "loss": 4.9042, "theoretical_loss": 5.583250282009159, "tokens_seen": 34865152 }, { "epoch": 0.01, "learning_rate": 0.0004997091273821465, "loss": 4.9647, "theoretical_loss": 5.581581577512031, "tokens_seen": 34930688 }, { "epoch": 0.01, "learning_rate": 0.0004996990972918757, "loss": 5.0026, "theoretical_loss": 5.579916875606134, "tokens_seen": 34996224 }, { "epoch": 0.01, "learning_rate": 0.0004996890672016048, "loss": 4.7549, "theoretical_loss": 5.578256159222196, "tokens_seen": 35061760 }, { "epoch": 0.01, "learning_rate": 0.000499679037111334, "loss": 4.6224, "theoretical_loss": 5.576599411395472, "tokens_seen": 35127296 }, { "epoch": 0.01, "learning_rate": 0.0004996690070210632, "loss": 4.7725, "theoretical_loss": 5.574946615264906, "tokens_seen": 35192832 }, { "epoch": 0.01, "learning_rate": 0.0004996589769307924, "loss": 4.7807, "theoretical_loss": 5.5732977540723105, "tokens_seen": 35258368 }, { "epoch": 0.01, "learning_rate": 0.0004996489468405216, "loss": 4.8423, "theoretical_loss": 5.571652811161542, "tokens_seen": 35323904 }, { "epoch": 0.01, "learning_rate": 0.0004996389167502507, "loss": 4.9602, "theoretical_loss": 5.570011769977693, "tokens_seen": 35389440 }, { "epoch": 0.01, "learning_rate": 0.0004996288866599799, "loss": 4.6312, "theoretical_loss": 5.568374614066299, "tokens_seen": 35454976 }, { "epoch": 0.01, "learning_rate": 0.0004996188565697091, "loss": 5.0662, "theoretical_loss": 5.566741327072535, "tokens_seen": 35520512 }, { "epoch": 0.01, "learning_rate": 0.0004996088264794383, "loss": 4.7143, "theoretical_loss": 5.565111892740433, "tokens_seen": 35586048 }, { "epoch": 0.01, "learning_rate": 0.0004995987963891675, "loss": 4.8642, "theoretical_loss": 5.563486294912105, "tokens_seen": 35651584 }, { "epoch": 0.01, "learning_rate": 0.0004995887662988968, "loss": 4.6962, "theoretical_loss": 5.56186451752697, "tokens_seen": 35717120 }, { "epoch": 0.01, "learning_rate": 0.0004995787362086258, "loss": 4.7386, "theoretical_loss": 5.560246544620993, "tokens_seen": 35782656 }, { "epoch": 0.01, "learning_rate": 0.0004995687061183551, "loss": 4.7368, "theoretical_loss": 5.558632360325929, "tokens_seen": 35848192 }, { "epoch": 0.01, "learning_rate": 0.0004995586760280842, "loss": 4.6722, "theoretical_loss": 5.557021948868571, "tokens_seen": 35913728 }, { "epoch": 0.01, "learning_rate": 0.0004995486459378135, "loss": 4.8891, "theoretical_loss": 5.555415294570011, "tokens_seen": 35979264 }, { "epoch": 0.01, "objective/train/docs_used": 32385, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.4478864669799805, "objective/train/theoretical_loss": 5.553812381844907, "objective/train/tokens_used": 56504800, "theoretical_loss": 5.553812381844907, "tokens_seen": 36044800 }, { "epoch": 0.01, "learning_rate": 0.0004995386158475427, "loss": 4.8836, "theoretical_loss": 5.553812381844907, "tokens_seen": 36044800 }, { "epoch": 0.01, "learning_rate": 0.0004995285857572718, "loss": 4.7132, "theoretical_loss": 5.552213195200755, "tokens_seen": 36110336 }, { "epoch": 0.01, "learning_rate": 0.000499518555667001, "loss": 4.7041, "theoretical_loss": 5.550617719237167, "tokens_seen": 36175872 }, { "epoch": 0.01, "learning_rate": 0.0004995085255767302, "loss": 4.8574, "theoretical_loss": 5.549025938645155, "tokens_seen": 36241408 }, { "epoch": 0.01, "learning_rate": 0.0004994984954864594, "loss": 4.743, "theoretical_loss": 5.547437838206435, "tokens_seen": 36306944 }, { "epoch": 0.01, "learning_rate": 0.0004994884653961886, "loss": 4.7443, "theoretical_loss": 5.545853402792717, "tokens_seen": 36372480 }, { "epoch": 0.01, "learning_rate": 0.0004994784353059178, "loss": 4.8769, "theoretical_loss": 5.544272617365014, "tokens_seen": 36438016 }, { "epoch": 0.01, "learning_rate": 0.0004994684052156469, "loss": 4.8343, "theoretical_loss": 5.542695466972956, "tokens_seen": 36503552 }, { "epoch": 0.01, "learning_rate": 0.0004994583751253761, "loss": 4.7822, "theoretical_loss": 5.541121936754111, "tokens_seen": 36569088 }, { "epoch": 0.01, "learning_rate": 0.0004994483450351053, "loss": 4.7453, "theoretical_loss": 5.539552011933312, "tokens_seen": 36634624 }, { "epoch": 0.01, "learning_rate": 0.0004994383149448345, "loss": 4.8192, "theoretical_loss": 5.537985677821986, "tokens_seen": 36700160 }, { "epoch": 0.01, "learning_rate": 0.0004994282848545637, "loss": 4.7944, "theoretical_loss": 5.536422919817495, "tokens_seen": 36765696 }, { "epoch": 0.01, "learning_rate": 0.0004994182547642928, "loss": 4.8368, "theoretical_loss": 5.5348637234024824, "tokens_seen": 36831232 }, { "epoch": 0.01, "learning_rate": 0.0004994082246740221, "loss": 4.8909, "theoretical_loss": 5.53330807414422, "tokens_seen": 36896768 }, { "epoch": 0.01, "learning_rate": 0.0004993981945837512, "loss": 4.7606, "theoretical_loss": 5.5317559576939725, "tokens_seen": 36962304 }, { "epoch": 0.01, "learning_rate": 0.0004993881644934805, "loss": 4.3933, "theoretical_loss": 5.530207359786353, "tokens_seen": 37027840 }, { "epoch": 0.01, "learning_rate": 0.0004993781344032096, "loss": 4.9105, "theoretical_loss": 5.5286622662386975, "tokens_seen": 37093376 }, { "epoch": 0.01, "learning_rate": 0.0004993681043129389, "loss": 4.725, "theoretical_loss": 5.52712066295044, "tokens_seen": 37158912 }, { "epoch": 0.01, "learning_rate": 0.000499358074222668, "loss": 4.6152, "theoretical_loss": 5.525582535902489, "tokens_seen": 37224448 }, { "epoch": 0.01, "learning_rate": 0.0004993480441323972, "loss": 4.8967, "theoretical_loss": 5.524047871156618, "tokens_seen": 37289984 }, { "epoch": 0.01, "learning_rate": 0.0004993380140421264, "loss": 4.304, "theoretical_loss": 5.52251665485486, "tokens_seen": 37355520 }, { "epoch": 0.01, "learning_rate": 0.0004993279839518556, "loss": 4.8395, "theoretical_loss": 5.520988873218897, "tokens_seen": 37421056 }, { "epoch": 0.01, "learning_rate": 0.0004993179538615848, "loss": 4.7291, "theoretical_loss": 5.519464512549478, "tokens_seen": 37486592 }, { "epoch": 0.01, "learning_rate": 0.0004993079237713139, "loss": 4.5346, "theoretical_loss": 5.5179435592258095, "tokens_seen": 37552128 }, { "epoch": 0.01, "learning_rate": 0.0004992978936810431, "loss": 4.7017, "theoretical_loss": 5.516425999704987, "tokens_seen": 37617664 }, { "epoch": 0.01, "objective/train/docs_used": 32883, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.716377258300781, "objective/train/theoretical_loss": 5.514911820521407, "objective/train/tokens_used": 58143200, "theoretical_loss": 5.514911820521407, "tokens_seen": 37683200 }, { "epoch": 0.01, "learning_rate": 0.0004992878635907723, "loss": 4.9009, "theoretical_loss": 5.514911820521407, "tokens_seen": 37683200 }, { "epoch": 0.01, "learning_rate": 0.0004992778335005015, "loss": 4.9884, "theoretical_loss": 5.5134010082861895, "tokens_seen": 37748736 }, { "epoch": 0.01, "learning_rate": 0.0004992678034102307, "loss": 4.7545, "theoretical_loss": 5.511893549686616, "tokens_seen": 37814272 }, { "epoch": 0.01, "learning_rate": 0.0004992577733199598, "loss": 4.7888, "theoretical_loss": 5.51038943148556, "tokens_seen": 37879808 }, { "epoch": 0.01, "learning_rate": 0.000499247743229689, "loss": 5.0533, "theoretical_loss": 5.508888640520928, "tokens_seen": 37945344 }, { "epoch": 0.01, "learning_rate": 0.0004992377131394183, "loss": 4.8807, "theoretical_loss": 5.50739116370511, "tokens_seen": 38010880 }, { "epoch": 0.01, "learning_rate": 0.0004992276830491475, "loss": 4.8592, "theoretical_loss": 5.505896988024423, "tokens_seen": 38076416 }, { "epoch": 0.01, "learning_rate": 0.0004992176529588767, "loss": 4.7331, "theoretical_loss": 5.5044061005385725, "tokens_seen": 38141952 }, { "epoch": 0.01, "learning_rate": 0.0004992076228686059, "loss": 4.8333, "theoretical_loss": 5.502918488380116, "tokens_seen": 38207488 }, { "epoch": 0.01, "learning_rate": 0.000499197592778335, "loss": 4.7589, "theoretical_loss": 5.501434138753918, "tokens_seen": 38273024 }, { "epoch": 0.01, "learning_rate": 0.0004991875626880642, "loss": 4.7598, "theoretical_loss": 5.499953038936635, "tokens_seen": 38338560 }, { "epoch": 0.01, "learning_rate": 0.0004991775325977934, "loss": 4.7997, "theoretical_loss": 5.498475176276176, "tokens_seen": 38404096 }, { "epoch": 0.01, "learning_rate": 0.0004991675025075226, "loss": 4.7163, "theoretical_loss": 5.497000538191195, "tokens_seen": 38469632 }, { "epoch": 0.01, "learning_rate": 0.0004991574724172518, "loss": 4.7824, "theoretical_loss": 5.495529112170568, "tokens_seen": 38535168 }, { "epoch": 0.01, "learning_rate": 0.000499147442326981, "loss": 4.8341, "theoretical_loss": 5.494060885772887, "tokens_seen": 38600704 }, { "epoch": 0.01, "learning_rate": 0.0004991374122367101, "loss": 4.7344, "theoretical_loss": 5.492595846625951, "tokens_seen": 38666240 }, { "epoch": 0.01, "learning_rate": 0.0004991273821464393, "loss": 4.7977, "theoretical_loss": 5.491133982426266, "tokens_seen": 38731776 }, { "epoch": 0.01, "learning_rate": 0.0004991173520561685, "loss": 4.3782, "theoretical_loss": 5.489675280938547, "tokens_seen": 38797312 }, { "epoch": 0.01, "learning_rate": 0.0004991073219658977, "loss": 4.7085, "theoretical_loss": 5.488219729995227, "tokens_seen": 38862848 }, { "epoch": 0.01, "learning_rate": 0.000499097291875627, "loss": 4.5818, "theoretical_loss": 5.486767317495966, "tokens_seen": 38928384 }, { "epoch": 0.01, "learning_rate": 0.000499087261785356, "loss": 5.0102, "theoretical_loss": 5.48531803140717, "tokens_seen": 38993920 }, { "epoch": 0.01, "learning_rate": 0.0004990772316950853, "loss": 4.6955, "theoretical_loss": 5.483871859761511, "tokens_seen": 39059456 }, { "epoch": 0.01, "learning_rate": 0.0004990672016048144, "loss": 4.6598, "theoretical_loss": 5.482428790657449, "tokens_seen": 39124992 }, { "epoch": 0.01, "learning_rate": 0.0004990571715145437, "loss": 4.822, "theoretical_loss": 5.480988812258763, "tokens_seen": 39190528 }, { "epoch": 0.01, "learning_rate": 0.0004990471414242729, "loss": 4.8752, "theoretical_loss": 5.479551912794086, "tokens_seen": 39256064 }, { "epoch": 0.01, "objective/train/docs_used": 34166, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.898425579071045, "objective/train/theoretical_loss": 5.478118080556438, "objective/train/tokens_used": 59781600, "theoretical_loss": 5.478118080556438, "tokens_seen": 39321600 }, { "epoch": 0.01, "learning_rate": 0.000499037111334002, "loss": 4.7202, "theoretical_loss": 5.478118080556438, "tokens_seen": 39321600 }, { "epoch": 0.01, "learning_rate": 0.0004990270812437312, "loss": 4.6432, "theoretical_loss": 5.476687303902768, "tokens_seen": 39387136 }, { "epoch": 0.01, "learning_rate": 0.0004990170511534604, "loss": 4.8362, "theoretical_loss": 5.475259571253502, "tokens_seen": 39452672 }, { "epoch": 0.01, "learning_rate": 0.0004990070210631896, "loss": 4.5375, "theoretical_loss": 5.473834871092089, "tokens_seen": 39518208 }, { "epoch": 0.01, "learning_rate": 0.0004989969909729188, "loss": 4.6575, "theoretical_loss": 5.4724131919645576, "tokens_seen": 39583744 }, { "epoch": 0.01, "learning_rate": 0.000498986960882648, "loss": 4.6854, "theoretical_loss": 5.470994522479069, "tokens_seen": 39649280 }, { "epoch": 0.01, "learning_rate": 0.0004989769307923771, "loss": 4.5501, "theoretical_loss": 5.4695788513054815, "tokens_seen": 39714816 }, { "epoch": 0.01, "learning_rate": 0.0004989669007021063, "loss": 4.7545, "theoretical_loss": 5.468166167174912, "tokens_seen": 39780352 }, { "epoch": 0.01, "learning_rate": 0.0004989568706118355, "loss": 4.714, "theoretical_loss": 5.466756458879306, "tokens_seen": 39845888 }, { "epoch": 0.01, "learning_rate": 0.0004989468405215647, "loss": 4.5583, "theoretical_loss": 5.465349715271013, "tokens_seen": 39911424 }, { "epoch": 0.01, "learning_rate": 0.0004989368104312939, "loss": 4.7862, "theoretical_loss": 5.463945925262355, "tokens_seen": 39976960 }, { "epoch": 0.01, "learning_rate": 0.000498926780341023, "loss": 4.4681, "theoretical_loss": 5.462545077825214, "tokens_seen": 40042496 }, { "epoch": 0.01, "learning_rate": 0.0004989167502507523, "loss": 4.9631, "theoretical_loss": 5.461147161990611, "tokens_seen": 40108032 }, { "epoch": 0.01, "learning_rate": 0.0004989067201604814, "loss": 4.662, "theoretical_loss": 5.459752166848292, "tokens_seen": 40173568 }, { "epoch": 0.01, "learning_rate": 0.0004988966900702107, "loss": 4.7581, "theoretical_loss": 5.458360081546321, "tokens_seen": 40239104 }, { "epoch": 0.01, "learning_rate": 0.0004988866599799398, "loss": 4.448, "theoretical_loss": 5.456970895290674, "tokens_seen": 40304640 }, { "epoch": 0.01, "learning_rate": 0.0004988766298896691, "loss": 4.548, "theoretical_loss": 5.455584597344835, "tokens_seen": 40370176 }, { "epoch": 0.01, "learning_rate": 0.0004988665997993982, "loss": 4.5585, "theoretical_loss": 5.454201177029395, "tokens_seen": 40435712 }, { "epoch": 0.01, "learning_rate": 0.0004988565697091274, "loss": 4.6426, "theoretical_loss": 5.452820623721662, "tokens_seen": 40501248 }, { "epoch": 0.01, "learning_rate": 0.0004988465396188566, "loss": 4.5398, "theoretical_loss": 5.45144292685526, "tokens_seen": 40566784 }, { "epoch": 0.01, "learning_rate": 0.0004988365095285858, "loss": 4.5747, "theoretical_loss": 5.450068075919752, "tokens_seen": 40632320 }, { "epoch": 0.01, "learning_rate": 0.000498826479438315, "loss": 4.6306, "theoretical_loss": 5.44869606046024, "tokens_seen": 40697856 }, { "epoch": 0.01, "learning_rate": 0.0004988164493480441, "loss": 4.6076, "theoretical_loss": 5.447326870076996, "tokens_seen": 40763392 }, { "epoch": 0.01, "learning_rate": 0.0004988064192577733, "loss": 4.6846, "theoretical_loss": 5.445960494425072, "tokens_seen": 40828928 }, { "epoch": 0.01, "learning_rate": 0.0004987963891675025, "loss": 4.5425, "theoretical_loss": 5.444596923213931, "tokens_seen": 40894464 }, { "epoch": 0.01, "objective/train/docs_used": 34866, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.677554607391357, "objective/train/theoretical_loss": 5.443236146207074, "objective/train/tokens_used": 61420000, "theoretical_loss": 5.443236146207074, "tokens_seen": 40960000 }, { "epoch": 0.01, "learning_rate": 0.0004987863590772317, "loss": 4.517, "theoretical_loss": 5.443236146207074, "tokens_seen": 40960000 }, { "epoch": 0.01, "learning_rate": 0.0004987763289869609, "loss": 4.4186, "theoretical_loss": 5.441878153221662, "tokens_seen": 41025536 }, { "epoch": 0.01, "learning_rate": 0.00049876629889669, "loss": 4.3927, "theoretical_loss": 5.440522934128164, "tokens_seen": 41091072 }, { "epoch": 0.01, "learning_rate": 0.0004987562688064192, "loss": 4.521, "theoretical_loss": 5.439170478849976, "tokens_seen": 41156608 }, { "epoch": 0.01, "learning_rate": 0.0004987462387161484, "loss": 4.4519, "theoretical_loss": 5.437820777363078, "tokens_seen": 41222144 }, { "epoch": 0.01, "learning_rate": 0.0004987362086258777, "loss": 4.5882, "theoretical_loss": 5.4364738196956655, "tokens_seen": 41287680 }, { "epoch": 0.01, "learning_rate": 0.0004987261785356068, "loss": 4.4504, "theoretical_loss": 5.435129595927794, "tokens_seen": 41353216 }, { "epoch": 0.01, "learning_rate": 0.0004987161484453361, "loss": 4.1807, "theoretical_loss": 5.433788096191039, "tokens_seen": 41418752 }, { "epoch": 0.01, "learning_rate": 0.0004987061183550651, "loss": 4.5904, "theoretical_loss": 5.432449310668134, "tokens_seen": 41484288 }, { "epoch": 0.01, "learning_rate": 0.0004986960882647944, "loss": 4.5925, "theoretical_loss": 5.4311132295926345, "tokens_seen": 41549824 }, { "epoch": 0.01, "learning_rate": 0.0004986860581745236, "loss": 4.6165, "theoretical_loss": 5.42977984324857, "tokens_seen": 41615360 }, { "epoch": 0.01, "learning_rate": 0.0004986760280842528, "loss": 4.49, "theoretical_loss": 5.428449141970107, "tokens_seen": 41680896 }, { "epoch": 0.01, "learning_rate": 0.000498665997993982, "loss": 4.5819, "theoretical_loss": 5.427121116141212, "tokens_seen": 41746432 }, { "epoch": 0.01, "learning_rate": 0.0004986559679037111, "loss": 4.7141, "theoretical_loss": 5.42579575619531, "tokens_seen": 41811968 }, { "epoch": 0.01, "learning_rate": 0.0004986459378134403, "loss": 4.6584, "theoretical_loss": 5.424473052614967, "tokens_seen": 41877504 }, { "epoch": 0.01, "learning_rate": 0.0004986359077231695, "loss": 4.7053, "theoretical_loss": 5.423152995931552, "tokens_seen": 41943040 }, { "epoch": 0.01, "learning_rate": 0.0004986258776328987, "loss": 4.677, "theoretical_loss": 5.421835576724906, "tokens_seen": 42008576 }, { "epoch": 0.01, "learning_rate": 0.0004986158475426279, "loss": 4.5363, "theoretical_loss": 5.420520785623031, "tokens_seen": 42074112 }, { "epoch": 0.01, "learning_rate": 0.000498605817452357, "loss": 4.4347, "theoretical_loss": 5.4192086133017625, "tokens_seen": 42139648 }, { "epoch": 0.01, "learning_rate": 0.0004985957873620862, "loss": 4.5436, "theoretical_loss": 5.417899050484451, "tokens_seen": 42205184 }, { "epoch": 0.01, "learning_rate": 0.0004985857572718154, "loss": 4.5196, "theoretical_loss": 5.416592087941646, "tokens_seen": 42270720 }, { "epoch": 0.01, "learning_rate": 0.0004985757271815446, "loss": 4.474, "theoretical_loss": 5.415287716490787, "tokens_seen": 42336256 }, { "epoch": 0.01, "learning_rate": 0.0004985656970912738, "loss": 4.9763, "theoretical_loss": 5.413985926995892, "tokens_seen": 42401792 }, { "epoch": 0.01, "learning_rate": 0.0004985556670010031, "loss": 4.4879, "theoretical_loss": 5.412686710367245, "tokens_seen": 42467328 }, { "epoch": 0.01, "learning_rate": 0.0004985456369107321, "loss": 4.7709, "theoretical_loss": 5.411390057561097, "tokens_seen": 42532864 }, { "epoch": 0.01, "objective/train/docs_used": 35868, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.189578056335449, "objective/train/theoretical_loss": 5.410095959579362, "objective/train/tokens_used": 63058400, "theoretical_loss": 5.410095959579362, "tokens_seen": 42598400 }, { "epoch": 0.01, "learning_rate": 0.0004985356068204614, "loss": 4.5885, "theoretical_loss": 5.410095959579362, "tokens_seen": 42598400 }, { "epoch": 0.01, "learning_rate": 0.0004985255767301905, "loss": 4.4367, "theoretical_loss": 5.408804407469308, "tokens_seen": 42663936 }, { "epoch": 0.01, "learning_rate": 0.0004985155466399198, "loss": 4.697, "theoretical_loss": 5.407515392323276, "tokens_seen": 42729472 }, { "epoch": 0.01, "learning_rate": 0.000498505516549649, "loss": 4.6402, "theoretical_loss": 5.406228905278368, "tokens_seen": 42795008 }, { "epoch": 0.01, "learning_rate": 0.0004984954864593782, "loss": 4.743, "theoretical_loss": 5.404944937516161, "tokens_seen": 42860544 }, { "epoch": 0.01, "learning_rate": 0.0004984854563691073, "loss": 4.4767, "theoretical_loss": 5.403663480262418, "tokens_seen": 42926080 }, { "epoch": 0.01, "learning_rate": 0.0004984754262788365, "loss": 4.2795, "theoretical_loss": 5.402384524786797, "tokens_seen": 42991616 }, { "epoch": 0.01, "learning_rate": 0.0004984653961885657, "loss": 4.5048, "theoretical_loss": 5.401108062402562, "tokens_seen": 43057152 }, { "epoch": 0.01, "learning_rate": 0.0004984553660982949, "loss": 4.285, "theoretical_loss": 5.399834084466306, "tokens_seen": 43122688 }, { "epoch": 0.01, "learning_rate": 0.0004984453360080241, "loss": 4.5244, "theoretical_loss": 5.398562582377666, "tokens_seen": 43188224 }, { "epoch": 0.01, "learning_rate": 0.0004984353059177532, "loss": 4.4912, "theoretical_loss": 5.397293547579041, "tokens_seen": 43253760 }, { "epoch": 0.01, "learning_rate": 0.0004984252758274825, "loss": 4.5286, "theoretical_loss": 5.396026971555319, "tokens_seen": 43319296 }, { "epoch": 0.01, "learning_rate": 0.0004984152457372116, "loss": 4.635, "theoretical_loss": 5.394762845833601, "tokens_seen": 43384832 }, { "epoch": 0.01, "learning_rate": 0.0004984052156469409, "loss": 4.6563, "theoretical_loss": 5.393501161982926, "tokens_seen": 43450368 }, { "epoch": 0.01, "learning_rate": 0.00049839518555667, "loss": 4.5338, "theoretical_loss": 5.392241911614005, "tokens_seen": 43515904 }, { "epoch": 0.01, "learning_rate": 0.0004983851554663993, "loss": 4.6497, "theoretical_loss": 5.390985086378949, "tokens_seen": 43581440 }, { "epoch": 0.01, "learning_rate": 0.0004983751253761284, "loss": 4.656, "theoretical_loss": 5.389730677971002, "tokens_seen": 43646976 }, { "epoch": 0.01, "learning_rate": 0.0004983650952858576, "loss": 4.2885, "theoretical_loss": 5.388478678124285, "tokens_seen": 43712512 }, { "epoch": 0.01, "learning_rate": 0.0004983550651955868, "loss": 4.6653, "theoretical_loss": 5.387229078613521, "tokens_seen": 43778048 }, { "epoch": 0.01, "learning_rate": 0.000498345035105316, "loss": 4.8537, "theoretical_loss": 5.385981871253785, "tokens_seen": 43843584 }, { "epoch": 0.01, "learning_rate": 0.0004983350050150452, "loss": 4.5315, "theoretical_loss": 5.384737047900243, "tokens_seen": 43909120 }, { "epoch": 0.01, "learning_rate": 0.0004983249749247743, "loss": 4.4477, "theoretical_loss": 5.3834946004478965, "tokens_seen": 43974656 }, { "epoch": 0.01, "learning_rate": 0.0004983149448345035, "loss": 4.2702, "theoretical_loss": 5.382254520831328, "tokens_seen": 44040192 }, { "epoch": 0.01, "learning_rate": 0.0004983049147442327, "loss": 4.4734, "theoretical_loss": 5.381016801024449, "tokens_seen": 44105728 }, { "epoch": 0.01, "learning_rate": 0.0004982948846539619, "loss": 4.5673, "theoretical_loss": 5.379781433040252, "tokens_seen": 44171264 }, { "epoch": 0.01, "objective/train/docs_used": 36544, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.2153215408325195, "objective/train/theoretical_loss": 5.378548408930558, "objective/train/tokens_used": 64696800, "theoretical_loss": 5.378548408930558, "tokens_seen": 44236800 }, { "epoch": 0.01, "learning_rate": 0.0004982848545636911, "loss": 4.3754, "theoretical_loss": 5.378548408930558, "tokens_seen": 44236800 }, { "epoch": 0.01, "learning_rate": 0.0004982748244734202, "loss": 4.6866, "theoretical_loss": 5.377317720785777, "tokens_seen": 44302336 }, { "epoch": 0.01, "learning_rate": 0.0004982647943831494, "loss": 4.2689, "theoretical_loss": 5.37608936073466, "tokens_seen": 44367872 }, { "epoch": 0.01, "learning_rate": 0.0004982547642928786, "loss": 4.558, "theoretical_loss": 5.374863320944057, "tokens_seen": 44433408 }, { "epoch": 0.01, "learning_rate": 0.0004982447342026079, "loss": 4.3621, "theoretical_loss": 5.373639593618675, "tokens_seen": 44498944 }, { "epoch": 0.01, "learning_rate": 0.000498234704112337, "loss": 4.7059, "theoretical_loss": 5.372418171000847, "tokens_seen": 44564480 }, { "epoch": 0.01, "learning_rate": 0.0004982246740220663, "loss": 4.4685, "theoretical_loss": 5.371199045370283, "tokens_seen": 44630016 }, { "epoch": 0.01, "learning_rate": 0.0004982146439317953, "loss": 4.502, "theoretical_loss": 5.369982209043851, "tokens_seen": 44695552 }, { "epoch": 0.01, "learning_rate": 0.0004982046138415246, "loss": 4.5419, "theoretical_loss": 5.368767654375327, "tokens_seen": 44761088 }, { "epoch": 0.01, "learning_rate": 0.0004981945837512538, "loss": 4.6487, "theoretical_loss": 5.367555373755179, "tokens_seen": 44826624 }, { "epoch": 0.01, "learning_rate": 0.000498184553660983, "loss": 4.5139, "theoretical_loss": 5.366345359610327, "tokens_seen": 44892160 }, { "epoch": 0.01, "learning_rate": 0.0004981745235707122, "loss": 4.6486, "theoretical_loss": 5.365137604403923, "tokens_seen": 44957696 }, { "epoch": 0.01, "learning_rate": 0.0004981644934804413, "loss": 4.6411, "theoretical_loss": 5.363932100635117, "tokens_seen": 45023232 }, { "epoch": 0.01, "learning_rate": 0.0004981544633901705, "loss": 4.4685, "theoretical_loss": 5.362728840838843, "tokens_seen": 45088768 }, { "epoch": 0.01, "learning_rate": 0.0004981444332998997, "loss": 4.3304, "theoretical_loss": 5.361527817585586, "tokens_seen": 45154304 }, { "epoch": 0.01, "learning_rate": 0.0004981344032096289, "loss": 4.5782, "theoretical_loss": 5.360329023481169, "tokens_seen": 45219840 }, { "epoch": 0.01, "learning_rate": 0.0004981243731193581, "loss": 4.4319, "theoretical_loss": 5.359132451166534, "tokens_seen": 45285376 }, { "epoch": 0.01, "learning_rate": 0.0004981143430290873, "loss": 4.6029, "theoretical_loss": 5.357938093317518, "tokens_seen": 45350912 }, { "epoch": 0.01, "learning_rate": 0.0004981043129388164, "loss": 4.458, "theoretical_loss": 5.356745942644645, "tokens_seen": 45416448 }, { "epoch": 0.01, "learning_rate": 0.0004980942828485456, "loss": 4.3535, "theoretical_loss": 5.355555991892905, "tokens_seen": 45481984 }, { "epoch": 0.01, "learning_rate": 0.0004980842527582748, "loss": 4.826, "theoretical_loss": 5.35436823384155, "tokens_seen": 45547520 }, { "epoch": 0.01, "learning_rate": 0.000498074222668004, "loss": 4.6619, "theoretical_loss": 5.353182661303873, "tokens_seen": 45613056 }, { "epoch": 0.01, "learning_rate": 0.0004980641925777333, "loss": 4.3435, "theoretical_loss": 5.35199926712701, "tokens_seen": 45678592 }, { "epoch": 0.01, "learning_rate": 0.0004980541624874623, "loss": 4.7139, "theoretical_loss": 5.350818044191721, "tokens_seen": 45744128 }, { "epoch": 0.01, "learning_rate": 0.0004980441323971916, "loss": 4.5414, "theoretical_loss": 5.349638985412193, "tokens_seen": 45809664 }, { "epoch": 0.01, "objective/train/docs_used": 37915, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.255738258361816, "objective/train/theoretical_loss": 5.348462083735834, "objective/train/tokens_used": 66335200, "theoretical_loss": 5.348462083735834, "tokens_seen": 45875200 }, { "epoch": 0.01, "learning_rate": 0.0004980341023069207, "loss": 4.2427, "theoretical_loss": 5.348462083735834, "tokens_seen": 45875200 }, { "epoch": 0.01, "learning_rate": 0.00049802407221665, "loss": 4.4766, "theoretical_loss": 5.347287332143064, "tokens_seen": 45940736 }, { "epoch": 0.01, "learning_rate": 0.0004980140421263792, "loss": 4.6487, "theoretical_loss": 5.346114723647119, "tokens_seen": 46006272 }, { "epoch": 0.01, "learning_rate": 0.0004980040120361084, "loss": 4.744, "theoretical_loss": 5.344944251293852, "tokens_seen": 46071808 }, { "epoch": 0.01, "learning_rate": 0.0004979939819458375, "loss": 4.4822, "theoretical_loss": 5.343775908161532, "tokens_seen": 46137344 }, { "epoch": 0.01, "learning_rate": 0.0004979839518555667, "loss": 4.6656, "theoretical_loss": 5.342609687360644, "tokens_seen": 46202880 }, { "epoch": 0.01, "learning_rate": 0.0004979739217652959, "loss": 4.368, "theoretical_loss": 5.341445582033705, "tokens_seen": 46268416 }, { "epoch": 0.01, "learning_rate": 0.0004979638916750251, "loss": 4.5362, "theoretical_loss": 5.3402835853550545, "tokens_seen": 46333952 }, { "epoch": 0.01, "learning_rate": 0.0004979538615847543, "loss": 4.5253, "theoretical_loss": 5.339123690530673, "tokens_seen": 46399488 }, { "epoch": 0.01, "learning_rate": 0.0004979438314944834, "loss": 4.5891, "theoretical_loss": 5.337965890797989, "tokens_seen": 46465024 }, { "epoch": 0.01, "learning_rate": 0.0004979338014042126, "loss": 4.3895, "theoretical_loss": 5.336810179425685, "tokens_seen": 46530560 }, { "epoch": 0.01, "learning_rate": 0.0004979237713139418, "loss": 4.7045, "theoretical_loss": 5.335656549713516, "tokens_seen": 46596096 }, { "epoch": 0.01, "learning_rate": 0.000497913741223671, "loss": 4.3343, "theoretical_loss": 5.334504994992115, "tokens_seen": 46661632 }, { "epoch": 0.01, "learning_rate": 0.0004979037111334002, "loss": 4.5711, "theoretical_loss": 5.333355508622814, "tokens_seen": 46727168 }, { "epoch": 0.01, "learning_rate": 0.0004978936810431293, "loss": 4.5964, "theoretical_loss": 5.332208083997459, "tokens_seen": 46792704 }, { "epoch": 0.01, "learning_rate": 0.0004978836509528586, "loss": 4.3615, "theoretical_loss": 5.33106271453822, "tokens_seen": 46858240 }, { "epoch": 0.01, "learning_rate": 0.0004978736208625877, "loss": 4.6588, "theoretical_loss": 5.329919393697422, "tokens_seen": 46923776 }, { "epoch": 0.01, "learning_rate": 0.000497863590772317, "loss": 4.8373, "theoretical_loss": 5.328778114957351, "tokens_seen": 46989312 }, { "epoch": 0.01, "learning_rate": 0.0004978535606820461, "loss": 4.2761, "theoretical_loss": 5.327638871830089, "tokens_seen": 47054848 }, { "epoch": 0.01, "learning_rate": 0.0004978435305917754, "loss": 4.4197, "theoretical_loss": 5.326501657857326, "tokens_seen": 47120384 }, { "epoch": 0.01, "learning_rate": 0.0004978335005015045, "loss": 4.435, "theoretical_loss": 5.32536646661019, "tokens_seen": 47185920 }, { "epoch": 0.01, "learning_rate": 0.0004978234704112337, "loss": 4.6838, "theoretical_loss": 5.324233291689069, "tokens_seen": 47251456 }, { "epoch": 0.01, "learning_rate": 0.0004978134403209629, "loss": 4.3913, "theoretical_loss": 5.323102126723439, "tokens_seen": 47316992 }, { "epoch": 0.01, "learning_rate": 0.0004978034102306921, "loss": 4.6, "theoretical_loss": 5.321972965371691, "tokens_seen": 47382528 }, { "epoch": 0.01, "learning_rate": 0.0004977933801404213, "loss": 4.1382, "theoretical_loss": 5.320845801320959, "tokens_seen": 47448064 }, { "epoch": 0.01, "objective/train/docs_used": 38651, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.045684814453125, "objective/train/theoretical_loss": 5.319720628286955, "objective/train/tokens_used": 67973600, "theoretical_loss": 5.319720628286955, "tokens_seen": 47513600 }, { "epoch": 0.01, "learning_rate": 0.0004977833500501504, "loss": 4.343, "theoretical_loss": 5.319720628286955, "tokens_seen": 47513600 }, { "epoch": 0.01, "learning_rate": 0.0004977733199598796, "loss": 4.3051, "theoretical_loss": 5.318597440013795, "tokens_seen": 47579136 }, { "epoch": 0.01, "learning_rate": 0.0004977632898696088, "loss": 4.6262, "theoretical_loss": 5.317476230273831, "tokens_seen": 47644672 }, { "epoch": 0.01, "learning_rate": 0.000497753259779338, "loss": 4.5661, "theoretical_loss": 5.316356992867491, "tokens_seen": 47710208 }, { "epoch": 0.01, "learning_rate": 0.0004977432296890672, "loss": 4.5954, "theoretical_loss": 5.31523972162311, "tokens_seen": 47775744 }, { "epoch": 0.01, "learning_rate": 0.0004977331995987965, "loss": 4.4414, "theoretical_loss": 5.314124410396767, "tokens_seen": 47841280 }, { "epoch": 0.01, "learning_rate": 0.0004977231695085255, "loss": 4.7758, "theoretical_loss": 5.31301105307212, "tokens_seen": 47906816 }, { "epoch": 0.01, "learning_rate": 0.0004977131394182548, "loss": 4.603, "theoretical_loss": 5.311899643560251, "tokens_seen": 47972352 }, { "epoch": 0.01, "learning_rate": 0.000497703109327984, "loss": 4.6245, "theoretical_loss": 5.310790175799497, "tokens_seen": 48037888 }, { "epoch": 0.01, "learning_rate": 0.0004976930792377132, "loss": 4.5712, "theoretical_loss": 5.3096826437553, "tokens_seen": 48103424 }, { "epoch": 0.01, "learning_rate": 0.0004976830491474424, "loss": 4.5459, "theoretical_loss": 5.308577041420046, "tokens_seen": 48168960 }, { "epoch": 0.01, "learning_rate": 0.0004976730190571715, "loss": 4.6043, "theoretical_loss": 5.3074733628129005, "tokens_seen": 48234496 }, { "epoch": 0.01, "learning_rate": 0.0004976629889669007, "loss": 4.5158, "theoretical_loss": 5.3063716019796665, "tokens_seen": 48300032 }, { "epoch": 0.01, "learning_rate": 0.0004976529588766299, "loss": 4.4244, "theoretical_loss": 5.305271752992619, "tokens_seen": 48365568 }, { "epoch": 0.01, "learning_rate": 0.0004976429287863591, "loss": 4.5599, "theoretical_loss": 5.304173809950358, "tokens_seen": 48431104 }, { "epoch": 0.01, "learning_rate": 0.0004976328986960883, "loss": 4.5297, "theoretical_loss": 5.303077766977653, "tokens_seen": 48496640 }, { "epoch": 0.01, "learning_rate": 0.0004976228686058175, "loss": 4.3215, "theoretical_loss": 5.3019836182252895, "tokens_seen": 48562176 }, { "epoch": 0.01, "learning_rate": 0.0004976128385155466, "loss": 4.3245, "theoretical_loss": 5.300891357869929, "tokens_seen": 48627712 }, { "epoch": 0.01, "learning_rate": 0.0004976028084252758, "loss": 4.2442, "theoretical_loss": 5.299800980113945, "tokens_seen": 48693248 }, { "epoch": 0.01, "learning_rate": 0.000497592778335005, "loss": 4.2681, "theoretical_loss": 5.298712479185288, "tokens_seen": 48758784 }, { "epoch": 0.01, "learning_rate": 0.0004975827482447342, "loss": 4.5281, "theoretical_loss": 5.297625849337331, "tokens_seen": 48824320 }, { "epoch": 0.01, "learning_rate": 0.0004975727181544635, "loss": 4.5414, "theoretical_loss": 5.296541084848727, "tokens_seen": 48889856 }, { "epoch": 0.01, "learning_rate": 0.0004975626880641925, "loss": 4.3081, "theoretical_loss": 5.295458180023262, "tokens_seen": 48955392 }, { "epoch": 0.01, "learning_rate": 0.0004975526579739218, "loss": 4.1252, "theoretical_loss": 5.294377129189715, "tokens_seen": 49020928 }, { "epoch": 0.01, "learning_rate": 0.0004975426278836509, "loss": 4.3003, "theoretical_loss": 5.293297926701706, "tokens_seen": 49086464 }, { "epoch": 0.01, "objective/train/docs_used": 39958, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.954103946685791, "objective/train/theoretical_loss": 5.292220566937567, "objective/train/tokens_used": 69612000, "theoretical_loss": 5.292220566937567, "tokens_seen": 49152000 }, { "epoch": 0.01, "learning_rate": 0.0004975325977933802, "loss": 4.5056, "theoretical_loss": 5.292220566937567, "tokens_seen": 49152000 }, { "epoch": 0.01, "learning_rate": 0.0004975225677031094, "loss": 4.549, "theoretical_loss": 5.29114504430019, "tokens_seen": 49217536 }, { "epoch": 0.01, "learning_rate": 0.0004975125376128386, "loss": 4.2312, "theoretical_loss": 5.290071353216895, "tokens_seen": 49283072 }, { "epoch": 0.01, "learning_rate": 0.0004975025075225677, "loss": 4.4122, "theoretical_loss": 5.288999488139284, "tokens_seen": 49348608 }, { "epoch": 0.01, "learning_rate": 0.0004974924774322969, "loss": 4.4153, "theoretical_loss": 5.28792944354311, "tokens_seen": 49414144 }, { "epoch": 0.01, "learning_rate": 0.0004974824473420261, "loss": 4.5622, "theoretical_loss": 5.286861213928137, "tokens_seen": 49479680 }, { "epoch": 0.02, "learning_rate": 0.0004974724172517553, "loss": 4.4577, "theoretical_loss": 5.285794793817999, "tokens_seen": 49545216 }, { "epoch": 0.02, "learning_rate": 0.0004974623871614845, "loss": 4.3913, "theoretical_loss": 5.284730177760077, "tokens_seen": 49610752 }, { "epoch": 0.02, "learning_rate": 0.0004974523570712136, "loss": 4.0983, "theoretical_loss": 5.283667360325351, "tokens_seen": 49676288 }, { "epoch": 0.02, "learning_rate": 0.0004974423269809428, "loss": 4.3874, "theoretical_loss": 5.2826063361082785, "tokens_seen": 49741824 }, { "epoch": 0.02, "learning_rate": 0.000497432296890672, "loss": 4.4498, "theoretical_loss": 5.281547099726654, "tokens_seen": 49807360 }, { "epoch": 0.02, "learning_rate": 0.0004974222668004012, "loss": 4.2229, "theoretical_loss": 5.280489645821483, "tokens_seen": 49872896 }, { "epoch": 0.02, "learning_rate": 0.0004974122367101304, "loss": 4.4956, "theoretical_loss": 5.279433969056848, "tokens_seen": 49938432 }, { "epoch": 0.02, "learning_rate": 0.0004974022066198595, "loss": 4.2938, "theoretical_loss": 5.278380064119782, "tokens_seen": 50003968 }, { "epoch": 0.02, "learning_rate": 0.0004973921765295888, "loss": 4.2649, "theoretical_loss": 5.277327925720137, "tokens_seen": 50069504 }, { "epoch": 0.02, "learning_rate": 0.0004973821464393179, "loss": 4.6124, "theoretical_loss": 5.276277548590457, "tokens_seen": 50135040 }, { "epoch": 0.02, "learning_rate": 0.0004973721163490472, "loss": 4.4276, "theoretical_loss": 5.275228927485855, "tokens_seen": 50200576 }, { "epoch": 0.02, "learning_rate": 0.0004973620862587763, "loss": 4.4451, "theoretical_loss": 5.2741820571838804, "tokens_seen": 50266112 }, { "epoch": 0.02, "learning_rate": 0.0004973520561685056, "loss": 4.4044, "theoretical_loss": 5.273136932484399, "tokens_seen": 50331648 }, { "epoch": 0.02, "learning_rate": 0.0004973420260782347, "loss": 4.1472, "theoretical_loss": 5.272093548209467, "tokens_seen": 50397184 }, { "epoch": 0.02, "learning_rate": 0.0004973319959879639, "loss": 4.3286, "theoretical_loss": 5.271051899203207, "tokens_seen": 50462720 }, { "epoch": 0.02, "learning_rate": 0.0004973219658976931, "loss": 4.4202, "theoretical_loss": 5.270011980331685, "tokens_seen": 50528256 }, { "epoch": 0.02, "learning_rate": 0.0004973119358074223, "loss": 4.3812, "theoretical_loss": 5.268973786482794, "tokens_seen": 50593792 }, { "epoch": 0.02, "learning_rate": 0.0004973019057171515, "loss": 4.4977, "theoretical_loss": 5.267937312566123, "tokens_seen": 50659328 }, { "epoch": 0.02, "learning_rate": 0.0004972918756268806, "loss": 4.4049, "theoretical_loss": 5.266902553512847, "tokens_seen": 50724864 }, { "epoch": 0.02, "objective/train/docs_used": 40584, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.171596527099609, "objective/train/theoretical_loss": 5.265869504275602, "objective/train/tokens_used": 71250400, "theoretical_loss": 5.265869504275602, "tokens_seen": 50790400 }, { "epoch": 0.02, "learning_rate": 0.0004972818455366098, "loss": 4.5105, "theoretical_loss": 5.265869504275602, "tokens_seen": 50790400 }, { "epoch": 0.02, "learning_rate": 0.000497271815446339, "loss": 4.2898, "theoretical_loss": 5.264838159828369, "tokens_seen": 50855936 }, { "epoch": 0.02, "learning_rate": 0.0004972617853560682, "loss": 4.5284, "theoretical_loss": 5.263808515166355, "tokens_seen": 50921472 }, { "epoch": 0.02, "learning_rate": 0.0004972517552657974, "loss": 4.1863, "theoretical_loss": 5.262780565305875, "tokens_seen": 50987008 }, { "epoch": 0.02, "learning_rate": 0.0004972417251755266, "loss": 4.1704, "theoretical_loss": 5.261754305284241, "tokens_seen": 51052544 }, { "epoch": 0.02, "learning_rate": 0.0004972316950852557, "loss": 4.4493, "theoretical_loss": 5.260729730159641, "tokens_seen": 51118080 }, { "epoch": 0.02, "learning_rate": 0.0004972216649949849, "loss": 4.1237, "theoretical_loss": 5.259706835011027, "tokens_seen": 51183616 }, { "epoch": 0.02, "learning_rate": 0.0004972116349047142, "loss": 4.3399, "theoretical_loss": 5.2586856149380035, "tokens_seen": 51249152 }, { "epoch": 0.02, "learning_rate": 0.0004972016048144433, "loss": 4.0849, "theoretical_loss": 5.257666065060709, "tokens_seen": 51314688 }, { "epoch": 0.02, "learning_rate": 0.0004971915747241726, "loss": 4.4541, "theoretical_loss": 5.256648180519708, "tokens_seen": 51380224 }, { "epoch": 0.02, "learning_rate": 0.0004971815446339017, "loss": 4.353, "theoretical_loss": 5.255631956475881, "tokens_seen": 51445760 }, { "epoch": 0.02, "learning_rate": 0.0004971715145436309, "loss": 4.2266, "theoretical_loss": 5.25461738811031, "tokens_seen": 51511296 }, { "epoch": 0.02, "learning_rate": 0.0004971614844533601, "loss": 4.4598, "theoretical_loss": 5.25360447062417, "tokens_seen": 51576832 }, { "epoch": 0.02, "learning_rate": 0.0004971514543630893, "loss": 4.5701, "theoretical_loss": 5.252593199238619, "tokens_seen": 51642368 }, { "epoch": 0.02, "learning_rate": 0.0004971414242728185, "loss": 4.2901, "theoretical_loss": 5.2515835691946915, "tokens_seen": 51707904 }, { "epoch": 0.02, "learning_rate": 0.0004971313941825477, "loss": 4.3521, "theoretical_loss": 5.2505755757531904, "tokens_seen": 51773440 }, { "epoch": 0.02, "learning_rate": 0.0004971213640922768, "loss": 4.3861, "theoretical_loss": 5.24956921419458, "tokens_seen": 51838976 }, { "epoch": 0.02, "learning_rate": 0.000497111334002006, "loss": 4.3723, "theoretical_loss": 5.248564479818876, "tokens_seen": 51904512 }, { "epoch": 0.02, "learning_rate": 0.0004971013039117352, "loss": 4.1067, "theoretical_loss": 5.247561367945544, "tokens_seen": 51970048 }, { "epoch": 0.02, "learning_rate": 0.0004970912738214644, "loss": 4.3448, "theoretical_loss": 5.246559873913396, "tokens_seen": 52035584 }, { "epoch": 0.02, "learning_rate": 0.0004970812437311936, "loss": 4.2188, "theoretical_loss": 5.245559993080484, "tokens_seen": 52101120 }, { "epoch": 0.02, "learning_rate": 0.0004970712136409227, "loss": 4.0193, "theoretical_loss": 5.24456172082399, "tokens_seen": 52166656 }, { "epoch": 0.02, "learning_rate": 0.0004970611835506519, "loss": 4.537, "theoretical_loss": 5.243565052540136, "tokens_seen": 52232192 }, { "epoch": 0.02, "learning_rate": 0.0004970511534603811, "loss": 4.4592, "theoretical_loss": 5.242569983644074, "tokens_seen": 52297728 }, { "epoch": 0.02, "learning_rate": 0.0004970411233701103, "loss": 4.2549, "theoretical_loss": 5.241576509569784, "tokens_seen": 52363264 }, { "epoch": 0.02, "objective/train/docs_used": 41941, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.2740397453308105, "objective/train/theoretical_loss": 5.240584625769978, "objective/train/tokens_used": 72888800, "theoretical_loss": 5.240584625769978, "tokens_seen": 52428800 }, { "epoch": 0.02, "learning_rate": 0.0004970310932798396, "loss": 4.1928, "theoretical_loss": 5.240584625769978, "tokens_seen": 52428800 }, { "epoch": 0.02, "learning_rate": 0.0004970210631895686, "loss": 4.4434, "theoretical_loss": 5.239594327715992, "tokens_seen": 52494336 }, { "epoch": 0.02, "learning_rate": 0.0004970110330992979, "loss": 4.5166, "theoretical_loss": 5.238605610897698, "tokens_seen": 52559872 }, { "epoch": 0.02, "learning_rate": 0.0004970010030090271, "loss": 3.9728, "theoretical_loss": 5.237618470823394, "tokens_seen": 52625408 }, { "epoch": 0.02, "learning_rate": 0.0004969909729187563, "loss": 4.3201, "theoretical_loss": 5.2366329030197125, "tokens_seen": 52690944 }, { "epoch": 0.02, "learning_rate": 0.0004969809428284855, "loss": 4.236, "theoretical_loss": 5.235648903031521, "tokens_seen": 52756480 }, { "epoch": 0.02, "learning_rate": 0.0004969709127382147, "loss": 4.3479, "theoretical_loss": 5.2346664664218245, "tokens_seen": 52822016 }, { "epoch": 0.02, "learning_rate": 0.0004969608826479438, "loss": 4.3036, "theoretical_loss": 5.233685588771669, "tokens_seen": 52887552 }, { "epoch": 0.02, "learning_rate": 0.000496950852557673, "loss": 4.3347, "theoretical_loss": 5.232706265680049, "tokens_seen": 52953088 }, { "epoch": 0.02, "learning_rate": 0.0004969408224674022, "loss": 4.3565, "theoretical_loss": 5.231728492763811, "tokens_seen": 53018624 }, { "epoch": 0.02, "learning_rate": 0.0004969307923771314, "loss": 4.6295, "theoretical_loss": 5.230752265657554, "tokens_seen": 53084160 }, { "epoch": 0.02, "learning_rate": 0.0004969207622868606, "loss": 4.3937, "theoretical_loss": 5.229777580013545, "tokens_seen": 53149696 }, { "epoch": 0.02, "learning_rate": 0.0004969107321965897, "loss": 4.4336, "theoretical_loss": 5.228804431501619, "tokens_seen": 53215232 }, { "epoch": 0.02, "learning_rate": 0.000496900702106319, "loss": 4.2758, "theoretical_loss": 5.227832815809087, "tokens_seen": 53280768 }, { "epoch": 0.02, "learning_rate": 0.0004968906720160481, "loss": 4.1819, "theoretical_loss": 5.226862728640651, "tokens_seen": 53346304 }, { "epoch": 0.02, "learning_rate": 0.0004968806419257774, "loss": 4.365, "theoretical_loss": 5.2258941657183, "tokens_seen": 53411840 }, { "epoch": 0.02, "learning_rate": 0.0004968706118355065, "loss": 4.316, "theoretical_loss": 5.2249271227812315, "tokens_seen": 53477376 }, { "epoch": 0.02, "learning_rate": 0.0004968605817452358, "loss": 4.0596, "theoretical_loss": 5.223961595585755, "tokens_seen": 53542912 }, { "epoch": 0.02, "learning_rate": 0.0004968505516549649, "loss": 4.3552, "theoretical_loss": 5.222997579905204, "tokens_seen": 53608448 }, { "epoch": 0.02, "learning_rate": 0.0004968405215646941, "loss": 4.435, "theoretical_loss": 5.222035071529845, "tokens_seen": 53673984 }, { "epoch": 0.02, "learning_rate": 0.0004968304914744233, "loss": 4.4653, "theoretical_loss": 5.2210740662667945, "tokens_seen": 53739520 }, { "epoch": 0.02, "learning_rate": 0.0004968204613841525, "loss": 4.4802, "theoretical_loss": 5.220114559939923, "tokens_seen": 53805056 }, { "epoch": 0.02, "learning_rate": 0.0004968104312938817, "loss": 4.3459, "theoretical_loss": 5.219156548389775, "tokens_seen": 53870592 }, { "epoch": 0.02, "learning_rate": 0.0004968004012036108, "loss": 4.4126, "theoretical_loss": 5.218200027473481, "tokens_seen": 53936128 }, { "epoch": 0.02, "learning_rate": 0.00049679037111334, "loss": 3.9646, "theoretical_loss": 5.217244993064664, "tokens_seen": 54001664 }, { "epoch": 0.02, "objective/train/docs_used": 42708, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.427297115325928, "objective/train/theoretical_loss": 5.216291441053366, "objective/train/tokens_used": 74527200, "theoretical_loss": 5.216291441053366, "tokens_seen": 54067200 }, { "epoch": 0.02, "learning_rate": 0.0004967803410230692, "loss": 4.2018, "theoretical_loss": 5.216291441053366, "tokens_seen": 54067200 }, { "epoch": 0.02, "learning_rate": 0.0004967703109327984, "loss": 4.5827, "theoretical_loss": 5.215339367345955, "tokens_seen": 54132736 }, { "epoch": 0.02, "learning_rate": 0.0004967602808425276, "loss": 4.4737, "theoretical_loss": 5.214388767865036, "tokens_seen": 54198272 }, { "epoch": 0.02, "learning_rate": 0.0004967502507522568, "loss": 4.2806, "theoretical_loss": 5.2134396385493815, "tokens_seen": 54263808 }, { "epoch": 0.02, "learning_rate": 0.0004967402206619859, "loss": 4.5709, "theoretical_loss": 5.212491975353835, "tokens_seen": 54329344 }, { "epoch": 0.02, "learning_rate": 0.0004967301905717151, "loss": 4.2733, "theoretical_loss": 5.211545774249233, "tokens_seen": 54394880 }, { "epoch": 0.02, "learning_rate": 0.0004967201604814444, "loss": 4.1417, "theoretical_loss": 5.210601031222324, "tokens_seen": 54460416 }, { "epoch": 0.02, "learning_rate": 0.0004967101303911735, "loss": 4.3607, "theoretical_loss": 5.209657742275683, "tokens_seen": 54525952 }, { "epoch": 0.02, "learning_rate": 0.0004967001003009028, "loss": 4.3571, "theoretical_loss": 5.208715903427631, "tokens_seen": 54591488 }, { "epoch": 0.02, "learning_rate": 0.000496690070210632, "loss": 4.1888, "theoretical_loss": 5.207775510712159, "tokens_seen": 54657024 }, { "epoch": 0.02, "learning_rate": 0.0004966800401203611, "loss": 4.3009, "theoretical_loss": 5.2068365601788384, "tokens_seen": 54722560 }, { "epoch": 0.02, "learning_rate": 0.0004966700100300903, "loss": 4.217, "theoretical_loss": 5.205899047892753, "tokens_seen": 54788096 }, { "epoch": 0.02, "learning_rate": 0.0004966599799398195, "loss": 4.2894, "theoretical_loss": 5.2049629699344075, "tokens_seen": 54853632 }, { "epoch": 0.02, "learning_rate": 0.0004966499498495487, "loss": 4.1144, "theoretical_loss": 5.204028322399658, "tokens_seen": 54919168 }, { "epoch": 0.02, "learning_rate": 0.0004966399197592779, "loss": 4.0488, "theoretical_loss": 5.203095101399628, "tokens_seen": 54984704 }, { "epoch": 0.02, "learning_rate": 0.000496629889669007, "loss": 4.272, "theoretical_loss": 5.202163303060633, "tokens_seen": 55050240 }, { "epoch": 0.02, "learning_rate": 0.0004966198595787362, "loss": 4.055, "theoretical_loss": 5.201232923524104, "tokens_seen": 55115776 }, { "epoch": 0.02, "learning_rate": 0.0004966098294884654, "loss": 4.5155, "theoretical_loss": 5.20030395894651, "tokens_seen": 55181312 }, { "epoch": 0.02, "learning_rate": 0.0004965997993981946, "loss": 4.467, "theoretical_loss": 5.199376405499277, "tokens_seen": 55246848 }, { "epoch": 0.02, "learning_rate": 0.0004965897693079238, "loss": 4.1735, "theoretical_loss": 5.198450259368721, "tokens_seen": 55312384 }, { "epoch": 0.02, "learning_rate": 0.0004965797392176529, "loss": 4.4801, "theoretical_loss": 5.197525516755965, "tokens_seen": 55377920 }, { "epoch": 0.02, "learning_rate": 0.0004965697091273821, "loss": 4.0844, "theoretical_loss": 5.196602173876867, "tokens_seen": 55443456 }, { "epoch": 0.02, "learning_rate": 0.0004965596790371113, "loss": 4.2085, "theoretical_loss": 5.195680226961947, "tokens_seen": 55508992 }, { "epoch": 0.02, "learning_rate": 0.0004965496489468405, "loss": 4.1159, "theoretical_loss": 5.194759672256309, "tokens_seen": 55574528 }, { "epoch": 0.02, "learning_rate": 0.0004965396188565698, "loss": 4.256, "theoretical_loss": 5.19384050601957, "tokens_seen": 55640064 }, { "epoch": 0.02, "objective/train/docs_used": 44116, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.755551815032959, "objective/train/theoretical_loss": 5.192922724525789, "objective/train/tokens_used": 76165600, "theoretical_loss": 5.192922724525789, "tokens_seen": 55705600 }, { "epoch": 0.02, "learning_rate": 0.0004965295887662988, "loss": 4.1988, "theoretical_loss": 5.192922724525789, "tokens_seen": 55705600 }, { "epoch": 0.02, "learning_rate": 0.0004965195586760281, "loss": 4.4217, "theoretical_loss": 5.19200632406339, "tokens_seen": 55771136 }, { "epoch": 0.02, "learning_rate": 0.0004965095285857573, "loss": 4.2675, "theoretical_loss": 5.19109130093509, "tokens_seen": 55836672 }, { "epoch": 0.02, "learning_rate": 0.0004964994984954865, "loss": 4.1937, "theoretical_loss": 5.190177651457833, "tokens_seen": 55902208 }, { "epoch": 0.02, "learning_rate": 0.0004964894684052157, "loss": 4.4199, "theoretical_loss": 5.189265371962712, "tokens_seen": 55967744 }, { "epoch": 0.02, "learning_rate": 0.0004964794383149449, "loss": 3.8862, "theoretical_loss": 5.188354458794902, "tokens_seen": 56033280 }, { "epoch": 0.02, "learning_rate": 0.000496469408224674, "loss": 4.3709, "theoretical_loss": 5.187444908313586, "tokens_seen": 56098816 }, { "epoch": 0.02, "learning_rate": 0.0004964593781344032, "loss": 3.9035, "theoretical_loss": 5.186536716891892, "tokens_seen": 56164352 }, { "epoch": 0.02, "learning_rate": 0.0004964493480441324, "loss": 4.1296, "theoretical_loss": 5.185629880916814, "tokens_seen": 56229888 }, { "epoch": 0.02, "learning_rate": 0.0004964393179538616, "loss": 4.0278, "theoretical_loss": 5.18472439678915, "tokens_seen": 56295424 }, { "epoch": 0.02, "learning_rate": 0.0004964292878635908, "loss": 4.1566, "theoretical_loss": 5.18382026092343, "tokens_seen": 56360960 }, { "epoch": 0.02, "learning_rate": 0.00049641925777332, "loss": 4.2582, "theoretical_loss": 5.182917469747851, "tokens_seen": 56426496 }, { "epoch": 0.02, "learning_rate": 0.0004964092276830491, "loss": 3.9274, "theoretical_loss": 5.182016019704204, "tokens_seen": 56492032 }, { "epoch": 0.02, "learning_rate": 0.0004963991975927783, "loss": 4.1733, "theoretical_loss": 5.1811159072478095, "tokens_seen": 56557568 }, { "epoch": 0.02, "learning_rate": 0.0004963891675025075, "loss": 4.1092, "theoretical_loss": 5.180217128847451, "tokens_seen": 56623104 }, { "epoch": 0.02, "learning_rate": 0.0004963791374122367, "loss": 3.9516, "theoretical_loss": 5.17931968098531, "tokens_seen": 56688640 }, { "epoch": 0.02, "learning_rate": 0.0004963691073219659, "loss": 4.113, "theoretical_loss": 5.178423560156894, "tokens_seen": 56754176 }, { "epoch": 0.02, "learning_rate": 0.0004963590772316951, "loss": 4.232, "theoretical_loss": 5.177528762870973, "tokens_seen": 56819712 }, { "epoch": 0.02, "learning_rate": 0.0004963490471414242, "loss": 4.2272, "theoretical_loss": 5.176635285649521, "tokens_seen": 56885248 }, { "epoch": 0.02, "learning_rate": 0.0004963390170511535, "loss": 4.1245, "theoretical_loss": 5.175743125027638, "tokens_seen": 56950784 }, { "epoch": 0.02, "learning_rate": 0.0004963289869608827, "loss": 4.2108, "theoretical_loss": 5.174852277553498, "tokens_seen": 57016320 }, { "epoch": 0.02, "learning_rate": 0.0004963189568706119, "loss": 4.4959, "theoretical_loss": 5.173962739788276, "tokens_seen": 57081856 }, { "epoch": 0.02, "learning_rate": 0.000496308926780341, "loss": 4.3544, "theoretical_loss": 5.17307450830609, "tokens_seen": 57147392 }, { "epoch": 0.02, "learning_rate": 0.0004962988966900702, "loss": 4.1915, "theoretical_loss": 5.172187579693933, "tokens_seen": 57212928 }, { "epoch": 0.02, "learning_rate": 0.0004962888665997994, "loss": 4.1823, "theoretical_loss": 5.1713019505516105, "tokens_seen": 57278464 }, { "epoch": 0.02, "objective/train/docs_used": 44698, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.967449188232422, "objective/train/theoretical_loss": 5.170417617491682, "objective/train/tokens_used": 77804000, "theoretical_loss": 5.170417617491682, "tokens_seen": 57344000 }, { "epoch": 0.02, "learning_rate": 0.0004962788365095286, "loss": 4.3458, "theoretical_loss": 5.170417617491682, "tokens_seen": 57344000 }, { "epoch": 0.02, "learning_rate": 0.0004962688064192578, "loss": 4.5212, "theoretical_loss": 5.169534577139395, "tokens_seen": 57409536 }, { "epoch": 0.02, "learning_rate": 0.000496258776328987, "loss": 4.2458, "theoretical_loss": 5.168652826132623, "tokens_seen": 57475072 }, { "epoch": 0.02, "learning_rate": 0.0004962487462387161, "loss": 4.1007, "theoretical_loss": 5.167772361121805, "tokens_seen": 57540608 }, { "epoch": 0.02, "learning_rate": 0.0004962387161484453, "loss": 4.2059, "theoretical_loss": 5.166893178769884, "tokens_seen": 57606144 }, { "epoch": 0.02, "learning_rate": 0.0004962286860581746, "loss": 4.1405, "theoretical_loss": 5.1660152757522475, "tokens_seen": 57671680 }, { "epoch": 0.02, "learning_rate": 0.0004962186559679037, "loss": 4.0437, "theoretical_loss": 5.165138648756665, "tokens_seen": 57737216 }, { "epoch": 0.02, "learning_rate": 0.000496208625877633, "loss": 3.8546, "theoretical_loss": 5.164263294483226, "tokens_seen": 57802752 }, { "epoch": 0.02, "learning_rate": 0.0004961985957873621, "loss": 3.9834, "theoretical_loss": 5.163389209644287, "tokens_seen": 57868288 }, { "epoch": 0.02, "learning_rate": 0.0004961885656970913, "loss": 3.9333, "theoretical_loss": 5.162516390964408, "tokens_seen": 57933824 }, { "epoch": 0.02, "learning_rate": 0.0004961785356068205, "loss": 4.0711, "theoretical_loss": 5.1616448351802875, "tokens_seen": 57999360 }, { "epoch": 0.02, "learning_rate": 0.0004961685055165497, "loss": 4.2007, "theoretical_loss": 5.160774539040716, "tokens_seen": 58064896 }, { "epoch": 0.02, "learning_rate": 0.0004961584754262789, "loss": 4.1938, "theoretical_loss": 5.159905499306511, "tokens_seen": 58130432 }, { "epoch": 0.02, "learning_rate": 0.0004961484453360081, "loss": 4.0338, "theoretical_loss": 5.159037712750455, "tokens_seen": 58195968 }, { "epoch": 0.02, "learning_rate": 0.0004961384152457372, "loss": 4.1705, "theoretical_loss": 5.158171176157245, "tokens_seen": 58261504 }, { "epoch": 0.02, "learning_rate": 0.0004961283851554664, "loss": 4.2802, "theoretical_loss": 5.157305886323435, "tokens_seen": 58327040 }, { "epoch": 0.02, "learning_rate": 0.0004961183550651956, "loss": 4.2534, "theoretical_loss": 5.156441840057371, "tokens_seen": 58392576 }, { "epoch": 0.02, "learning_rate": 0.0004961083249749248, "loss": 4.072, "theoretical_loss": 5.155579034179144, "tokens_seen": 58458112 }, { "epoch": 0.02, "learning_rate": 0.000496098294884654, "loss": 4.2144, "theoretical_loss": 5.15471746552053, "tokens_seen": 58523648 }, { "epoch": 0.02, "learning_rate": 0.0004960882647943831, "loss": 4.1997, "theoretical_loss": 5.153857130924929, "tokens_seen": 58589184 }, { "epoch": 0.02, "learning_rate": 0.0004960782347041123, "loss": 4.2174, "theoretical_loss": 5.1529980272473175, "tokens_seen": 58654720 }, { "epoch": 0.02, "learning_rate": 0.0004960682046138415, "loss": 4.3491, "theoretical_loss": 5.152140151354191, "tokens_seen": 58720256 }, { "epoch": 0.02, "learning_rate": 0.0004960581745235707, "loss": 4.1085, "theoretical_loss": 5.151283500123505, "tokens_seen": 58785792 }, { "epoch": 0.02, "learning_rate": 0.0004960481444333, "loss": 4.1087, "theoretical_loss": 5.150428070444621, "tokens_seen": 58851328 }, { "epoch": 0.02, "learning_rate": 0.000496038114343029, "loss": 4.0075, "theoretical_loss": 5.149573859218261, "tokens_seen": 58916864 }, { "epoch": 0.02, "objective/train/docs_used": 46033, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.9614341259002686, "objective/train/theoretical_loss": 5.1487208633564405, "objective/train/tokens_used": 79442400, "theoretical_loss": 5.1487208633564405, "tokens_seen": 58982400 }, { "epoch": 0.02, "learning_rate": 0.0004960280842527583, "loss": 4.125, "theoretical_loss": 5.1487208633564405, "tokens_seen": 58982400 }, { "epoch": 0.02, "learning_rate": 0.0004960180541624875, "loss": 4.0149, "theoretical_loss": 5.147869079782423, "tokens_seen": 59047936 }, { "epoch": 0.02, "learning_rate": 0.0004960080240722167, "loss": 4.2106, "theoretical_loss": 5.147018505430666, "tokens_seen": 59113472 }, { "epoch": 0.02, "learning_rate": 0.0004959979939819459, "loss": 4.2445, "theoretical_loss": 5.146169137246765, "tokens_seen": 59179008 }, { "epoch": 0.02, "learning_rate": 0.0004959879638916751, "loss": 4.2846, "theoretical_loss": 5.145320972187402, "tokens_seen": 59244544 }, { "epoch": 0.02, "learning_rate": 0.0004959779338014042, "loss": 4.0537, "theoretical_loss": 5.144474007220293, "tokens_seen": 59310080 }, { "epoch": 0.02, "learning_rate": 0.0004959679037111334, "loss": 4.3207, "theoretical_loss": 5.143628239324139, "tokens_seen": 59375616 }, { "epoch": 0.02, "learning_rate": 0.0004959578736208626, "loss": 4.0034, "theoretical_loss": 5.142783665488567, "tokens_seen": 59441152 }, { "epoch": 0.02, "learning_rate": 0.0004959478435305918, "loss": 3.8394, "theoretical_loss": 5.1419402827140885, "tokens_seen": 59506688 }, { "epoch": 0.02, "learning_rate": 0.000495937813440321, "loss": 4.0548, "theoretical_loss": 5.141098088012036, "tokens_seen": 59572224 }, { "epoch": 0.02, "learning_rate": 0.0004959277833500501, "loss": 3.6982, "theoretical_loss": 5.140257078404524, "tokens_seen": 59637760 }, { "epoch": 0.02, "learning_rate": 0.0004959177532597793, "loss": 4.1571, "theoretical_loss": 5.13941725092439, "tokens_seen": 59703296 }, { "epoch": 0.02, "learning_rate": 0.0004959077231695085, "loss": 4.1842, "theoretical_loss": 5.138578602615146, "tokens_seen": 59768832 }, { "epoch": 0.02, "learning_rate": 0.0004958976930792377, "loss": 4.2649, "theoretical_loss": 5.137741130530934, "tokens_seen": 59834368 }, { "epoch": 0.02, "learning_rate": 0.0004958876629889669, "loss": 4.1585, "theoretical_loss": 5.1369048317364685, "tokens_seen": 59899904 }, { "epoch": 0.02, "learning_rate": 0.000495877632898696, "loss": 4.1038, "theoretical_loss": 5.13606970330699, "tokens_seen": 59965440 }, { "epoch": 0.02, "learning_rate": 0.0004958676028084253, "loss": 4.0452, "theoretical_loss": 5.135235742328217, "tokens_seen": 60030976 }, { "epoch": 0.02, "learning_rate": 0.0004958575727181544, "loss": 4.2665, "theoretical_loss": 5.134402945896297, "tokens_seen": 60096512 }, { "epoch": 0.02, "learning_rate": 0.0004958475426278837, "loss": 4.0482, "theoretical_loss": 5.133571311117755, "tokens_seen": 60162048 }, { "epoch": 0.02, "learning_rate": 0.0004958375125376129, "loss": 4.1642, "theoretical_loss": 5.132740835109448, "tokens_seen": 60227584 }, { "epoch": 0.02, "learning_rate": 0.0004958274824473421, "loss": 4.2918, "theoretical_loss": 5.131911514998518, "tokens_seen": 60293120 }, { "epoch": 0.02, "learning_rate": 0.0004958174523570712, "loss": 4.0217, "theoretical_loss": 5.131083347922338, "tokens_seen": 60358656 }, { "epoch": 0.02, "learning_rate": 0.0004958074222668004, "loss": 4.0733, "theoretical_loss": 5.130256331028474, "tokens_seen": 60424192 }, { "epoch": 0.02, "learning_rate": 0.0004957973921765296, "loss": 4.5194, "theoretical_loss": 5.129430461474628, "tokens_seen": 60489728 }, { "epoch": 0.02, "learning_rate": 0.0004957873620862588, "loss": 4.0056, "theoretical_loss": 5.128605736428597, "tokens_seen": 60555264 }, { "epoch": 0.02, "objective/train/docs_used": 46630, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.424596786499023, "objective/train/theoretical_loss": 5.127782153068225, "objective/train/tokens_used": 81080800, "theoretical_loss": 5.127782153068225, "tokens_seen": 60620800 }, { "epoch": 0.02, "learning_rate": 0.000495777331995988, "loss": 4.4411, "theoretical_loss": 5.127782153068225, "tokens_seen": 60620800 }, { "epoch": 0.02, "learning_rate": 0.0004957673019057172, "loss": 3.919, "theoretical_loss": 5.126959708581356, "tokens_seen": 60686336 }, { "epoch": 0.02, "learning_rate": 0.0004957572718154463, "loss": 4.032, "theoretical_loss": 5.1261384001657895, "tokens_seen": 60751872 }, { "epoch": 0.02, "learning_rate": 0.0004957472417251755, "loss": 4.1885, "theoretical_loss": 5.125318225029231, "tokens_seen": 60817408 }, { "epoch": 0.02, "learning_rate": 0.0004957372116349047, "loss": 4.0579, "theoretical_loss": 5.124499180389249, "tokens_seen": 60882944 }, { "epoch": 0.02, "learning_rate": 0.0004957271815446339, "loss": 3.8165, "theoretical_loss": 5.12368126347323, "tokens_seen": 60948480 }, { "epoch": 0.02, "learning_rate": 0.0004957171514543631, "loss": 3.8252, "theoretical_loss": 5.122864471518334, "tokens_seen": 61014016 }, { "epoch": 0.02, "learning_rate": 0.0004957071213640923, "loss": 4.0573, "theoretical_loss": 5.122048801771443, "tokens_seen": 61079552 }, { "epoch": 0.02, "learning_rate": 0.0004956970912738214, "loss": 4.0186, "theoretical_loss": 5.121234251489128, "tokens_seen": 61145088 }, { "epoch": 0.02, "learning_rate": 0.0004956870611835507, "loss": 4.0424, "theoretical_loss": 5.120420817937591, "tokens_seen": 61210624 }, { "epoch": 0.02, "learning_rate": 0.0004956770310932798, "loss": 4.2235, "theoretical_loss": 5.119608498392633, "tokens_seen": 61276160 }, { "epoch": 0.02, "learning_rate": 0.0004956670010030091, "loss": 3.7588, "theoretical_loss": 5.118797290139605, "tokens_seen": 61341696 }, { "epoch": 0.02, "learning_rate": 0.0004956569709127383, "loss": 4.0559, "theoretical_loss": 5.117987190473361, "tokens_seen": 61407232 }, { "epoch": 0.02, "learning_rate": 0.0004956469408224674, "loss": 4.3216, "theoretical_loss": 5.1171781966982195, "tokens_seen": 61472768 }, { "epoch": 0.02, "learning_rate": 0.0004956369107321966, "loss": 3.99, "theoretical_loss": 5.116370306127921, "tokens_seen": 61538304 }, { "epoch": 0.02, "learning_rate": 0.0004956268806419258, "loss": 3.96, "theoretical_loss": 5.11556351608558, "tokens_seen": 61603840 }, { "epoch": 0.02, "learning_rate": 0.000495616850551655, "loss": 3.9305, "theoretical_loss": 5.114757823903647, "tokens_seen": 61669376 }, { "epoch": 0.02, "learning_rate": 0.0004956068204613842, "loss": 3.885, "theoretical_loss": 5.113953226923864, "tokens_seen": 61734912 }, { "epoch": 0.02, "learning_rate": 0.0004955967903711133, "loss": 4.2558, "theoretical_loss": 5.113149722497221, "tokens_seen": 61800448 }, { "epoch": 0.02, "learning_rate": 0.0004955867602808425, "loss": 4.0224, "theoretical_loss": 5.112347307983919, "tokens_seen": 61865984 }, { "epoch": 0.02, "learning_rate": 0.0004955767301905717, "loss": 4.0078, "theoretical_loss": 5.111545980753322, "tokens_seen": 61931520 }, { "epoch": 0.02, "learning_rate": 0.0004955667001003009, "loss": 4.2075, "theoretical_loss": 5.110745738183919, "tokens_seen": 61997056 }, { "epoch": 0.02, "learning_rate": 0.0004955566700100301, "loss": 4.2822, "theoretical_loss": 5.109946577663284, "tokens_seen": 62062592 }, { "epoch": 0.02, "learning_rate": 0.0004955466399197592, "loss": 3.8552, "theoretical_loss": 5.109148496588032, "tokens_seen": 62128128 }, { "epoch": 0.02, "learning_rate": 0.0004955366098294884, "loss": 4.1157, "theoretical_loss": 5.108351492363779, "tokens_seen": 62193664 }, { "epoch": 0.02, "objective/train/docs_used": 47896, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.253730773925781, "objective/train/theoretical_loss": 5.107555562405102, "objective/train/tokens_used": 82719200, "theoretical_loss": 5.107555562405102, "tokens_seen": 62259200 }, { "epoch": 0.02, "learning_rate": 0.0004955265797392177, "loss": 3.8586, "theoretical_loss": 5.107555562405102, "tokens_seen": 62259200 }, { "epoch": 0.02, "learning_rate": 0.0004955165496489468, "loss": 4.043, "theoretical_loss": 5.106760704135499, "tokens_seen": 62324736 }, { "epoch": 0.02, "learning_rate": 0.0004955065195586761, "loss": 3.9028, "theoretical_loss": 5.105966914987349, "tokens_seen": 62390272 }, { "epoch": 0.02, "learning_rate": 0.0004954964894684052, "loss": 4.1462, "theoretical_loss": 5.1051741924018685, "tokens_seen": 62455808 }, { "epoch": 0.02, "learning_rate": 0.0004954864593781344, "loss": 3.9346, "theoretical_loss": 5.10438253382908, "tokens_seen": 62521344 }, { "epoch": 0.02, "learning_rate": 0.0004954764292878636, "loss": 4.1546, "theoretical_loss": 5.103591936727762, "tokens_seen": 62586880 }, { "epoch": 0.02, "learning_rate": 0.0004954663991975928, "loss": 4.1114, "theoretical_loss": 5.102802398565418, "tokens_seen": 62652416 }, { "epoch": 0.02, "learning_rate": 0.000495456369107322, "loss": 4.2456, "theoretical_loss": 5.102013916818235, "tokens_seen": 62717952 }, { "epoch": 0.02, "learning_rate": 0.0004954463390170512, "loss": 4.1117, "theoretical_loss": 5.101226488971042, "tokens_seen": 62783488 }, { "epoch": 0.02, "learning_rate": 0.0004954363089267803, "loss": 3.8726, "theoretical_loss": 5.100440112517276, "tokens_seen": 62849024 }, { "epoch": 0.02, "learning_rate": 0.0004954262788365095, "loss": 3.9981, "theoretical_loss": 5.09965478495894, "tokens_seen": 62914560 }, { "epoch": 0.02, "learning_rate": 0.0004954162487462387, "loss": 4.048, "theoretical_loss": 5.098870503806567, "tokens_seen": 62980096 }, { "epoch": 0.02, "learning_rate": 0.0004954062186559679, "loss": 3.8959, "theoretical_loss": 5.09808726657918, "tokens_seen": 63045632 }, { "epoch": 0.02, "learning_rate": 0.0004953961885656971, "loss": 3.7875, "theoretical_loss": 5.097305070804255, "tokens_seen": 63111168 }, { "epoch": 0.02, "learning_rate": 0.0004953861584754263, "loss": 4.108, "theoretical_loss": 5.096523914017688, "tokens_seen": 63176704 }, { "epoch": 0.02, "learning_rate": 0.0004953761283851555, "loss": 4.1691, "theoretical_loss": 5.095743793763747, "tokens_seen": 63242240 }, { "epoch": 0.02, "learning_rate": 0.0004953660982948846, "loss": 4.0009, "theoretical_loss": 5.094964707595047, "tokens_seen": 63307776 }, { "epoch": 0.02, "learning_rate": 0.0004953560682046139, "loss": 3.7485, "theoretical_loss": 5.094186653072505, "tokens_seen": 63373312 }, { "epoch": 0.02, "learning_rate": 0.0004953460381143431, "loss": 4.0331, "theoretical_loss": 5.093409627765306, "tokens_seen": 63438848 }, { "epoch": 0.02, "learning_rate": 0.0004953360080240723, "loss": 3.9678, "theoretical_loss": 5.092633629250866, "tokens_seen": 63504384 }, { "epoch": 0.02, "learning_rate": 0.0004953259779338014, "loss": 4.031, "theoretical_loss": 5.091858655114796, "tokens_seen": 63569920 }, { "epoch": 0.02, "learning_rate": 0.0004953159478435306, "loss": 4.0263, "theoretical_loss": 5.091084702950868, "tokens_seen": 63635456 }, { "epoch": 0.02, "learning_rate": 0.0004953059177532598, "loss": 3.9067, "theoretical_loss": 5.090311770360971, "tokens_seen": 63700992 }, { "epoch": 0.02, "learning_rate": 0.000495295887662989, "loss": 4.0007, "theoretical_loss": 5.089539854955088, "tokens_seen": 63766528 }, { "epoch": 0.02, "learning_rate": 0.0004952858575727182, "loss": 3.7926, "theoretical_loss": 5.088768954351249, "tokens_seen": 63832064 }, { "epoch": 0.02, "objective/train/docs_used": 48551, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.194922924041748, "objective/train/theoretical_loss": 5.087999066175502, "objective/train/tokens_used": 84357600, "theoretical_loss": 5.087999066175502, "tokens_seen": 63897600 }, { "epoch": 0.02, "learning_rate": 0.0004952758274824474, "loss": 4.0708, "theoretical_loss": 5.087999066175502, "tokens_seen": 63897600 }, { "epoch": 0.02, "learning_rate": 0.0004952657973921765, "loss": 4.289, "theoretical_loss": 5.0872301880618735, "tokens_seen": 63963136 }, { "epoch": 0.02, "learning_rate": 0.0004952557673019057, "loss": 4.0031, "theoretical_loss": 5.086462317652341, "tokens_seen": 64028672 }, { "epoch": 0.02, "learning_rate": 0.0004952457372116349, "loss": 3.7044, "theoretical_loss": 5.085695452596788, "tokens_seen": 64094208 }, { "epoch": 0.02, "learning_rate": 0.0004952357071213641, "loss": 3.8383, "theoretical_loss": 5.084929590552976, "tokens_seen": 64159744 }, { "epoch": 0.02, "learning_rate": 0.0004952256770310933, "loss": 3.8943, "theoretical_loss": 5.0841647291865115, "tokens_seen": 64225280 }, { "epoch": 0.02, "learning_rate": 0.0004952156469408225, "loss": 4.0131, "theoretical_loss": 5.083400866170806, "tokens_seen": 64290816 }, { "epoch": 0.02, "learning_rate": 0.0004952056168505516, "loss": 3.8292, "theoretical_loss": 5.082637999187046, "tokens_seen": 64356352 }, { "epoch": 0.02, "learning_rate": 0.0004951955867602809, "loss": 3.8947, "theoretical_loss": 5.081876125924159, "tokens_seen": 64421888 }, { "epoch": 0.02, "learning_rate": 0.00049518555667001, "loss": 3.8301, "theoretical_loss": 5.0811152440787755, "tokens_seen": 64487424 }, { "epoch": 0.02, "learning_rate": 0.0004951755265797393, "loss": 3.9172, "theoretical_loss": 5.0803553513552036, "tokens_seen": 64552960 }, { "epoch": 0.02, "learning_rate": 0.0004951654964894685, "loss": 3.9521, "theoretical_loss": 5.079596445465386, "tokens_seen": 64618496 }, { "epoch": 0.02, "learning_rate": 0.0004951554663991976, "loss": 3.874, "theoretical_loss": 5.078838524128878, "tokens_seen": 64684032 }, { "epoch": 0.02, "learning_rate": 0.0004951454363089268, "loss": 3.886, "theoretical_loss": 5.078081585072802, "tokens_seen": 64749568 }, { "epoch": 0.02, "learning_rate": 0.000495135406218656, "loss": 3.9186, "theoretical_loss": 5.077325626031826, "tokens_seen": 64815104 }, { "epoch": 0.02, "learning_rate": 0.0004951253761283852, "loss": 3.8648, "theoretical_loss": 5.076570644748123, "tokens_seen": 64880640 }, { "epoch": 0.02, "learning_rate": 0.0004951153460381144, "loss": 4.1708, "theoretical_loss": 5.075816638971341, "tokens_seen": 64946176 }, { "epoch": 0.02, "learning_rate": 0.0004951053159478435, "loss": 3.8125, "theoretical_loss": 5.075063606458576, "tokens_seen": 65011712 }, { "epoch": 0.02, "learning_rate": 0.0004950952858575727, "loss": 3.7121, "theoretical_loss": 5.074311544974331, "tokens_seen": 65077248 }, { "epoch": 0.02, "learning_rate": 0.0004950852557673019, "loss": 3.7576, "theoretical_loss": 5.07356045229049, "tokens_seen": 65142784 }, { "epoch": 0.02, "learning_rate": 0.0004950752256770311, "loss": 3.9499, "theoretical_loss": 5.072810326186285, "tokens_seen": 65208320 }, { "epoch": 0.02, "learning_rate": 0.0004950651955867603, "loss": 3.7499, "theoretical_loss": 5.072061164448261, "tokens_seen": 65273856 }, { "epoch": 0.02, "learning_rate": 0.0004950551654964894, "loss": 3.833, "theoretical_loss": 5.071312964870252, "tokens_seen": 65339392 }, { "epoch": 0.02, "learning_rate": 0.0004950451354062186, "loss": 4.1678, "theoretical_loss": 5.070565725253344, "tokens_seen": 65404928 }, { "epoch": 0.02, "learning_rate": 0.0004950351053159479, "loss": 3.7484, "theoretical_loss": 5.069819443405842, "tokens_seen": 65470464 }, { "epoch": 0.02, "objective/train/docs_used": 49128, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.299447536468506, "objective/train/theoretical_loss": 5.069074117143246, "objective/train/tokens_used": 85996000, "theoretical_loss": 5.069074117143246, "tokens_seen": 65536000 }, { "epoch": 0.02, "learning_rate": 0.000495025075225677, "loss": 4.2265, "theoretical_loss": 5.069074117143246, "tokens_seen": 65536000 }, { "epoch": 0.02, "learning_rate": 0.0004950150451354063, "loss": 3.8292, "theoretical_loss": 5.068329744288216, "tokens_seen": 65601536 }, { "epoch": 0.02, "learning_rate": 0.0004950050150451354, "loss": 3.9529, "theoretical_loss": 5.067586322670541, "tokens_seen": 65667072 }, { "epoch": 0.02, "learning_rate": 0.0004949949849548646, "loss": 3.9573, "theoretical_loss": 5.0668438501271105, "tokens_seen": 65732608 }, { "epoch": 0.02, "learning_rate": 0.0004949849548645938, "loss": 3.8014, "theoretical_loss": 5.066102324501883, "tokens_seen": 65798144 }, { "epoch": 0.02, "learning_rate": 0.000494974924774323, "loss": 4.0132, "theoretical_loss": 5.065361743645855, "tokens_seen": 65863680 }, { "epoch": 0.02, "learning_rate": 0.0004949648946840522, "loss": 3.8756, "theoretical_loss": 5.064622105417033, "tokens_seen": 65929216 }, { "epoch": 0.02, "learning_rate": 0.0004949548645937814, "loss": 3.938, "theoretical_loss": 5.063883407680405, "tokens_seen": 65994752 }, { "epoch": 0.02, "learning_rate": 0.0004949448345035105, "loss": 3.7738, "theoretical_loss": 5.063145648307904, "tokens_seen": 66060288 }, { "epoch": 0.02, "learning_rate": 0.0004949348044132397, "loss": 3.8514, "theoretical_loss": 5.062408825178388, "tokens_seen": 66125824 }, { "epoch": 0.02, "learning_rate": 0.0004949247743229689, "loss": 3.8407, "theoretical_loss": 5.061672936177604, "tokens_seen": 66191360 }, { "epoch": 0.02, "learning_rate": 0.0004949147442326981, "loss": 3.949, "theoretical_loss": 5.06093797919816, "tokens_seen": 66256896 }, { "epoch": 0.02, "learning_rate": 0.0004949047141424273, "loss": 3.9103, "theoretical_loss": 5.060203952139497, "tokens_seen": 66322432 }, { "epoch": 0.02, "learning_rate": 0.0004948946840521565, "loss": 3.7828, "theoretical_loss": 5.059470852907861, "tokens_seen": 66387968 }, { "epoch": 0.02, "learning_rate": 0.0004948846539618856, "loss": 4.0738, "theoretical_loss": 5.0587386794162725, "tokens_seen": 66453504 }, { "epoch": 0.02, "learning_rate": 0.0004948746238716148, "loss": 3.7987, "theoretical_loss": 5.058007429584498, "tokens_seen": 66519040 }, { "epoch": 0.02, "learning_rate": 0.000494864593781344, "loss": 4.0708, "theoretical_loss": 5.057277101339023, "tokens_seen": 66584576 }, { "epoch": 0.02, "learning_rate": 0.0004948545636910733, "loss": 3.9817, "theoretical_loss": 5.056547692613021, "tokens_seen": 66650112 }, { "epoch": 0.02, "learning_rate": 0.0004948445336008024, "loss": 3.9141, "theoretical_loss": 5.055819201346331, "tokens_seen": 66715648 }, { "epoch": 0.02, "learning_rate": 0.0004948345035105316, "loss": 3.7643, "theoretical_loss": 5.055091625485421, "tokens_seen": 66781184 }, { "epoch": 0.02, "learning_rate": 0.0004948244734202607, "loss": 3.9492, "theoretical_loss": 5.054364962983367, "tokens_seen": 66846720 }, { "epoch": 0.02, "learning_rate": 0.00049481444332999, "loss": 3.7609, "theoretical_loss": 5.053639211799824, "tokens_seen": 66912256 }, { "epoch": 0.02, "learning_rate": 0.0004948044132397192, "loss": 4.1554, "theoretical_loss": 5.052914369900997, "tokens_seen": 66977792 }, { "epoch": 0.02, "learning_rate": 0.0004947943831494484, "loss": 3.6762, "theoretical_loss": 5.052190435259614, "tokens_seen": 67043328 }, { "epoch": 0.02, "learning_rate": 0.0004947843530591776, "loss": 3.8481, "theoretical_loss": 5.051467405854897, "tokens_seen": 67108864 }, { "epoch": 0.02, "objective/train/docs_used": 50658, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4438347816467285, "objective/train/theoretical_loss": 5.05074527967254, "objective/train/tokens_used": 87634400, "theoretical_loss": 5.05074527967254, "tokens_seen": 67174400 }, { "epoch": 0.02, "learning_rate": 0.0004947743229689067, "loss": 3.7927, "theoretical_loss": 5.05074527967254, "tokens_seen": 67174400 }, { "epoch": 0.02, "learning_rate": 0.0004947642928786359, "loss": 3.9986, "theoretical_loss": 5.050024054704677, "tokens_seen": 67239936 }, { "epoch": 0.02, "learning_rate": 0.0004947542627883651, "loss": 3.8366, "theoretical_loss": 5.049303728949859, "tokens_seen": 67305472 }, { "epoch": 0.02, "learning_rate": 0.0004947442326980943, "loss": 3.6166, "theoretical_loss": 5.048584300413019, "tokens_seen": 67371008 }, { "epoch": 0.02, "learning_rate": 0.0004947342026078235, "loss": 3.666, "theoretical_loss": 5.04786576710546, "tokens_seen": 67436544 }, { "epoch": 0.02, "learning_rate": 0.0004947241725175527, "loss": 3.6312, "theoretical_loss": 5.0471481270448155, "tokens_seen": 67502080 }, { "epoch": 0.02, "learning_rate": 0.0004947141424272818, "loss": 3.6529, "theoretical_loss": 5.046431378255027, "tokens_seen": 67567616 }, { "epoch": 0.02, "learning_rate": 0.0004947041123370111, "loss": 4.1165, "theoretical_loss": 5.045715518766322, "tokens_seen": 67633152 }, { "epoch": 0.02, "learning_rate": 0.0004946940822467402, "loss": 3.9446, "theoretical_loss": 5.0450005466151815, "tokens_seen": 67698688 }, { "epoch": 0.02, "learning_rate": 0.0004946840521564695, "loss": 3.916, "theoretical_loss": 5.044286459844319, "tokens_seen": 67764224 }, { "epoch": 0.02, "learning_rate": 0.0004946740220661987, "loss": 4.0487, "theoretical_loss": 5.043573256502652, "tokens_seen": 67829760 }, { "epoch": 0.02, "learning_rate": 0.0004946639919759278, "loss": 3.751, "theoretical_loss": 5.0428609346452795, "tokens_seen": 67895296 }, { "epoch": 0.02, "learning_rate": 0.000494653961885657, "loss": 4.055, "theoretical_loss": 5.042149492333452, "tokens_seen": 67960832 }, { "epoch": 0.02, "learning_rate": 0.0004946439317953862, "loss": 3.8691, "theoretical_loss": 5.041438927634549, "tokens_seen": 68026368 }, { "epoch": 0.02, "learning_rate": 0.0004946339017051154, "loss": 3.9087, "theoretical_loss": 5.040729238622053, "tokens_seen": 68091904 }, { "epoch": 0.02, "learning_rate": 0.0004946238716148446, "loss": 3.9294, "theoretical_loss": 5.040020423375525, "tokens_seen": 68157440 }, { "epoch": 0.02, "learning_rate": 0.0004946138415245737, "loss": 3.8602, "theoretical_loss": 5.039312479980579, "tokens_seen": 68222976 }, { "epoch": 0.02, "learning_rate": 0.0004946038114343029, "loss": 3.9095, "theoretical_loss": 5.038605406528857, "tokens_seen": 68288512 }, { "epoch": 0.02, "learning_rate": 0.0004945937813440321, "loss": 4.2147, "theoretical_loss": 5.037899201118005, "tokens_seen": 68354048 }, { "epoch": 0.02, "learning_rate": 0.0004945837512537613, "loss": 3.6029, "theoretical_loss": 5.037193861851646, "tokens_seen": 68419584 }, { "epoch": 0.02, "learning_rate": 0.0004945737211634905, "loss": 3.8454, "theoretical_loss": 5.03648938683936, "tokens_seen": 68485120 }, { "epoch": 0.02, "learning_rate": 0.0004945636910732196, "loss": 3.9214, "theoretical_loss": 5.035785774196654, "tokens_seen": 68550656 }, { "epoch": 0.02, "learning_rate": 0.0004945536609829488, "loss": 3.6581, "theoretical_loss": 5.035083022044944, "tokens_seen": 68616192 }, { "epoch": 0.02, "learning_rate": 0.0004945436308926781, "loss": 3.6618, "theoretical_loss": 5.034381128511525, "tokens_seen": 68681728 }, { "epoch": 0.02, "learning_rate": 0.0004945336008024072, "loss": 3.7457, "theoretical_loss": 5.0336800917295506, "tokens_seen": 68747264 }, { "epoch": 0.02, "objective/train/docs_used": 51428, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.8635945320129395, "objective/train/theoretical_loss": 5.032979909838007, "objective/train/tokens_used": 89272800, "theoretical_loss": 5.032979909838007, "tokens_seen": 68812800 }, { "epoch": 0.02, "learning_rate": 0.0004945235707121365, "loss": 3.842, "theoretical_loss": 5.032979909838007, "tokens_seen": 68812800 }, { "epoch": 0.02, "learning_rate": 0.0004945135406218656, "loss": 4.0085, "theoretical_loss": 5.032280580981691, "tokens_seen": 68878336 }, { "epoch": 0.02, "learning_rate": 0.0004945035105315948, "loss": 3.8896, "theoretical_loss": 5.031582103311187, "tokens_seen": 68943872 }, { "epoch": 0.02, "learning_rate": 0.000494493480441324, "loss": 3.8326, "theoretical_loss": 5.030884474982842, "tokens_seen": 69009408 }, { "epoch": 0.02, "learning_rate": 0.0004944834503510532, "loss": 3.8144, "theoretical_loss": 5.030187694158739, "tokens_seen": 69074944 }, { "epoch": 0.02, "learning_rate": 0.0004944734202607824, "loss": 3.797, "theoretical_loss": 5.02949175900668, "tokens_seen": 69140480 }, { "epoch": 0.02, "learning_rate": 0.0004944633901705116, "loss": 3.9856, "theoretical_loss": 5.028796667700159, "tokens_seen": 69206016 }, { "epoch": 0.02, "learning_rate": 0.0004944533600802407, "loss": 3.8468, "theoretical_loss": 5.0281024184183405, "tokens_seen": 69271552 }, { "epoch": 0.02, "learning_rate": 0.0004944433299899699, "loss": 3.706, "theoretical_loss": 5.0274090093460355, "tokens_seen": 69337088 }, { "epoch": 0.02, "learning_rate": 0.0004944332998996991, "loss": 3.692, "theoretical_loss": 5.026716438673677, "tokens_seen": 69402624 }, { "epoch": 0.02, "learning_rate": 0.0004944232698094283, "loss": 3.9909, "theoretical_loss": 5.0260247045973045, "tokens_seen": 69468160 }, { "epoch": 0.02, "learning_rate": 0.0004944132397191575, "loss": 3.9516, "theoretical_loss": 5.02533380531853, "tokens_seen": 69533696 }, { "epoch": 0.02, "learning_rate": 0.0004944032096288867, "loss": 4.0669, "theoretical_loss": 5.024643739044526, "tokens_seen": 69599232 }, { "epoch": 0.02, "learning_rate": 0.0004943931795386158, "loss": 3.8696, "theoretical_loss": 5.023954503987998, "tokens_seen": 69664768 }, { "epoch": 0.02, "learning_rate": 0.000494383149448345, "loss": 3.8331, "theoretical_loss": 5.023266098367161, "tokens_seen": 69730304 }, { "epoch": 0.02, "learning_rate": 0.0004943731193580742, "loss": 3.9494, "theoretical_loss": 5.022578520405721, "tokens_seen": 69795840 }, { "epoch": 0.02, "learning_rate": 0.0004943630892678035, "loss": 3.4961, "theoretical_loss": 5.0218917683328534, "tokens_seen": 69861376 }, { "epoch": 0.02, "learning_rate": 0.0004943530591775326, "loss": 3.8677, "theoretical_loss": 5.021205840383175, "tokens_seen": 69926912 }, { "epoch": 0.02, "learning_rate": 0.0004943430290872618, "loss": 3.6711, "theoretical_loss": 5.020520734796728, "tokens_seen": 69992448 }, { "epoch": 0.02, "learning_rate": 0.0004943329989969909, "loss": 3.8103, "theoretical_loss": 5.019836449818957, "tokens_seen": 70057984 }, { "epoch": 0.02, "learning_rate": 0.0004943229689067202, "loss": 3.9293, "theoretical_loss": 5.019152983700687, "tokens_seen": 70123520 }, { "epoch": 0.02, "learning_rate": 0.0004943129388164494, "loss": 3.6354, "theoretical_loss": 5.018470334698101, "tokens_seen": 70189056 }, { "epoch": 0.02, "learning_rate": 0.0004943029087261786, "loss": 3.5086, "theoretical_loss": 5.01778850107272, "tokens_seen": 70254592 }, { "epoch": 0.02, "learning_rate": 0.0004942928786359078, "loss": 3.5231, "theoretical_loss": 5.017107481091379, "tokens_seen": 70320128 }, { "epoch": 0.02, "learning_rate": 0.0004942828485456369, "loss": 3.7571, "theoretical_loss": 5.016427273026212, "tokens_seen": 70385664 }, { "epoch": 0.02, "objective/train/docs_used": 52769, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7651798725128174, "objective/train/theoretical_loss": 5.015747875154622, "objective/train/tokens_used": 90911200, "theoretical_loss": 5.015747875154622, "tokens_seen": 70451200 }, { "epoch": 0.02, "learning_rate": 0.0004942728184553661, "loss": 3.8123, "theoretical_loss": 5.015747875154622, "tokens_seen": 70451200 }, { "epoch": 0.02, "learning_rate": 0.0004942627883650953, "loss": 3.593, "theoretical_loss": 5.015069285759269, "tokens_seen": 70516736 }, { "epoch": 0.02, "learning_rate": 0.0004942527582748245, "loss": 3.6413, "theoretical_loss": 5.01439150312804, "tokens_seen": 70582272 }, { "epoch": 0.02, "learning_rate": 0.0004942427281845537, "loss": 3.6802, "theoretical_loss": 5.0137145255540405, "tokens_seen": 70647808 }, { "epoch": 0.02, "learning_rate": 0.0004942326980942828, "loss": 3.6497, "theoretical_loss": 5.013038351335559, "tokens_seen": 70713344 }, { "epoch": 0.02, "learning_rate": 0.000494222668004012, "loss": 4.0304, "theoretical_loss": 5.012362978776057, "tokens_seen": 70778880 }, { "epoch": 0.02, "learning_rate": 0.0004942126379137412, "loss": 3.6552, "theoretical_loss": 5.011688406184147, "tokens_seen": 70844416 }, { "epoch": 0.02, "learning_rate": 0.0004942026078234704, "loss": 3.5983, "theoretical_loss": 5.011014631873566, "tokens_seen": 70909952 }, { "epoch": 0.02, "learning_rate": 0.0004941925777331996, "loss": 3.9619, "theoretical_loss": 5.010341654163167, "tokens_seen": 70975488 }, { "epoch": 0.02, "learning_rate": 0.0004941825476429289, "loss": 3.6699, "theoretical_loss": 5.009669471376882, "tokens_seen": 71041024 }, { "epoch": 0.02, "learning_rate": 0.0004941725175526579, "loss": 3.8407, "theoretical_loss": 5.008998081843721, "tokens_seen": 71106560 }, { "epoch": 0.02, "learning_rate": 0.0004941624874623872, "loss": 3.7259, "theoretical_loss": 5.008327483897736, "tokens_seen": 71172096 }, { "epoch": 0.02, "learning_rate": 0.0004941524573721163, "loss": 3.5894, "theoretical_loss": 5.00765767587801, "tokens_seen": 71237632 }, { "epoch": 0.02, "learning_rate": 0.0004941424272818456, "loss": 3.576, "theoretical_loss": 5.006988656128635, "tokens_seen": 71303168 }, { "epoch": 0.02, "learning_rate": 0.0004941323971915748, "loss": 3.9922, "theoretical_loss": 5.006320422998691, "tokens_seen": 71368704 }, { "epoch": 0.02, "learning_rate": 0.0004941223671013039, "loss": 3.8547, "theoretical_loss": 5.00565297484223, "tokens_seen": 71434240 }, { "epoch": 0.02, "learning_rate": 0.0004941123370110331, "loss": 3.8379, "theoretical_loss": 5.004986310018252, "tokens_seen": 71499776 }, { "epoch": 0.02, "learning_rate": 0.0004941023069207623, "loss": 3.7754, "theoretical_loss": 5.004320426890686, "tokens_seen": 71565312 }, { "epoch": 0.02, "learning_rate": 0.0004940922768304915, "loss": 3.6593, "theoretical_loss": 5.003655323828376, "tokens_seen": 71630848 }, { "epoch": 0.02, "learning_rate": 0.0004940822467402207, "loss": 3.6119, "theoretical_loss": 5.002990999205057, "tokens_seen": 71696384 }, { "epoch": 0.02, "learning_rate": 0.0004940722166499498, "loss": 3.9359, "theoretical_loss": 5.002327451399335, "tokens_seen": 71761920 }, { "epoch": 0.02, "learning_rate": 0.000494062186559679, "loss": 3.947, "theoretical_loss": 5.001664678794671, "tokens_seen": 71827456 }, { "epoch": 0.02, "learning_rate": 0.0004940521564694082, "loss": 4.1257, "theoretical_loss": 5.001002679779363, "tokens_seen": 71892992 }, { "epoch": 0.02, "learning_rate": 0.0004940421263791374, "loss": 3.5797, "theoretical_loss": 5.0003414527465235, "tokens_seen": 71958528 }, { "epoch": 0.02, "learning_rate": 0.0004940320962888666, "loss": 3.4931, "theoretical_loss": 4.99968099609406, "tokens_seen": 72024064 }, { "epoch": 0.02, "objective/train/docs_used": 53229, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.443143129348755, "objective/train/theoretical_loss": 4.999021308224664, "objective/train/tokens_used": 92549600, "theoretical_loss": 4.999021308224664, "tokens_seen": 72089600 }, { "epoch": 0.02, "learning_rate": 0.0004940220661985958, "loss": 3.8379, "theoretical_loss": 4.999021308224664, "tokens_seen": 72089600 }, { "epoch": 0.02, "learning_rate": 0.0004940120361083249, "loss": 3.7169, "theoretical_loss": 4.998362387545782, "tokens_seen": 72155136 }, { "epoch": 0.02, "learning_rate": 0.0004940020060180542, "loss": 3.7454, "theoretical_loss": 4.997704232469606, "tokens_seen": 72220672 }, { "epoch": 0.02, "learning_rate": 0.0004939919759277834, "loss": 3.7546, "theoretical_loss": 4.997046841413049, "tokens_seen": 72286208 }, { "epoch": 0.02, "learning_rate": 0.0004939819458375126, "loss": 3.7608, "theoretical_loss": 4.996390212797728, "tokens_seen": 72351744 }, { "epoch": 0.02, "learning_rate": 0.0004939719157472418, "loss": 3.9281, "theoretical_loss": 4.995734345049949, "tokens_seen": 72417280 }, { "epoch": 0.02, "learning_rate": 0.000493961885656971, "loss": 3.6721, "theoretical_loss": 4.995079236600686, "tokens_seen": 72482816 }, { "epoch": 0.02, "learning_rate": 0.0004939518555667001, "loss": 3.5009, "theoretical_loss": 4.994424885885564, "tokens_seen": 72548352 }, { "epoch": 0.02, "learning_rate": 0.0004939418254764293, "loss": 4.0443, "theoretical_loss": 4.993771291344839, "tokens_seen": 72613888 }, { "epoch": 0.02, "learning_rate": 0.0004939317953861585, "loss": 3.851, "theoretical_loss": 4.993118451423381, "tokens_seen": 72679424 }, { "epoch": 0.02, "learning_rate": 0.0004939217652958877, "loss": 3.8891, "theoretical_loss": 4.992466364570659, "tokens_seen": 72744960 }, { "epoch": 0.02, "learning_rate": 0.0004939117352056169, "loss": 3.7893, "theoretical_loss": 4.991815029240721, "tokens_seen": 72810496 }, { "epoch": 0.02, "learning_rate": 0.000493901705115346, "loss": 3.8364, "theoretical_loss": 4.991164443892175, "tokens_seen": 72876032 }, { "epoch": 0.02, "learning_rate": 0.0004938916750250752, "loss": 3.4055, "theoretical_loss": 4.990514606988173, "tokens_seen": 72941568 }, { "epoch": 0.02, "learning_rate": 0.0004938816449348044, "loss": 3.4979, "theoretical_loss": 4.989865516996396, "tokens_seen": 73007104 }, { "epoch": 0.02, "learning_rate": 0.0004938716148445337, "loss": 3.8367, "theoretical_loss": 4.98921717238903, "tokens_seen": 73072640 }, { "epoch": 0.02, "learning_rate": 0.0004938615847542628, "loss": 3.8357, "theoretical_loss": 4.988569571642756, "tokens_seen": 73138176 }, { "epoch": 0.02, "learning_rate": 0.000493851554663992, "loss": 3.7075, "theoretical_loss": 4.98792271323873, "tokens_seen": 73203712 }, { "epoch": 0.02, "learning_rate": 0.0004938415245737211, "loss": 3.939, "theoretical_loss": 4.9872765956625615, "tokens_seen": 73269248 }, { "epoch": 0.02, "learning_rate": 0.0004938314944834504, "loss": 3.8696, "theoretical_loss": 4.9866312174043035, "tokens_seen": 73334784 }, { "epoch": 0.02, "learning_rate": 0.0004938214643931796, "loss": 3.8818, "theoretical_loss": 4.9859865769584335, "tokens_seen": 73400320 }, { "epoch": 0.02, "learning_rate": 0.0004938114343029088, "loss": 3.8674, "theoretical_loss": 4.9853426728238315, "tokens_seen": 73465856 }, { "epoch": 0.02, "learning_rate": 0.000493801404212638, "loss": 3.7658, "theoretical_loss": 4.984699503503771, "tokens_seen": 73531392 }, { "epoch": 0.02, "learning_rate": 0.0004937913741223671, "loss": 3.7394, "theoretical_loss": 4.984057067505898, "tokens_seen": 73596928 }, { "epoch": 0.02, "learning_rate": 0.0004937813440320963, "loss": 3.8743, "theoretical_loss": 4.9834153633422105, "tokens_seen": 73662464 }, { "epoch": 0.02, "objective/train/docs_used": 53865, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.384488582611084, "objective/train/theoretical_loss": 4.982774389529053, "objective/train/tokens_used": 94188000, "theoretical_loss": 4.982774389529053, "tokens_seen": 73728000 }, { "epoch": 0.02, "learning_rate": 0.0004937713139418255, "loss": 4.1487, "theoretical_loss": 4.982774389529053, "tokens_seen": 73728000 }, { "epoch": 0.02, "learning_rate": 0.0004937612838515547, "loss": 3.535, "theoretical_loss": 4.9821341445870875, "tokens_seen": 73793536 }, { "epoch": 0.02, "learning_rate": 0.0004937512537612839, "loss": 3.769, "theoretical_loss": 4.981494627041286, "tokens_seen": 73859072 }, { "epoch": 0.02, "learning_rate": 0.000493741223671013, "loss": 3.6317, "theoretical_loss": 4.98085583542091, "tokens_seen": 73924608 }, { "epoch": 0.02, "learning_rate": 0.0004937311935807422, "loss": 3.9226, "theoretical_loss": 4.980217768259496, "tokens_seen": 73990144 }, { "epoch": 0.02, "learning_rate": 0.0004937211634904714, "loss": 3.7611, "theoretical_loss": 4.979580424094836, "tokens_seen": 74055680 }, { "epoch": 0.02, "learning_rate": 0.0004937111334002006, "loss": 3.765, "theoretical_loss": 4.978943801468967, "tokens_seen": 74121216 }, { "epoch": 0.02, "learning_rate": 0.0004937011033099298, "loss": 3.6153, "theoretical_loss": 4.978307898928149, "tokens_seen": 74186752 }, { "epoch": 0.02, "learning_rate": 0.0004936910732196591, "loss": 3.7389, "theoretical_loss": 4.977672715022855, "tokens_seen": 74252288 }, { "epoch": 0.02, "learning_rate": 0.0004936810431293881, "loss": 3.5885, "theoretical_loss": 4.97703824830775, "tokens_seen": 74317824 }, { "epoch": 0.02, "learning_rate": 0.0004936710130391174, "loss": 3.7634, "theoretical_loss": 4.976404497341676, "tokens_seen": 74383360 }, { "epoch": 0.02, "learning_rate": 0.0004936609829488465, "loss": 3.62, "theoretical_loss": 4.975771460687641, "tokens_seen": 74448896 }, { "epoch": 0.02, "learning_rate": 0.0004936509528585758, "loss": 3.8546, "theoretical_loss": 4.975139136912794, "tokens_seen": 74514432 }, { "epoch": 0.02, "learning_rate": 0.000493640922768305, "loss": 3.768, "theoretical_loss": 4.974507524588424, "tokens_seen": 74579968 }, { "epoch": 0.02, "learning_rate": 0.0004936308926780341, "loss": 3.9096, "theoretical_loss": 4.973876622289927, "tokens_seen": 74645504 }, { "epoch": 0.02, "learning_rate": 0.0004936208625877633, "loss": 3.6973, "theoretical_loss": 4.973246428596802, "tokens_seen": 74711040 }, { "epoch": 0.02, "learning_rate": 0.0004936108324974925, "loss": 3.672, "theoretical_loss": 4.972616942092634, "tokens_seen": 74776576 }, { "epoch": 0.02, "learning_rate": 0.0004936008024072217, "loss": 3.4177, "theoretical_loss": 4.971988161365077, "tokens_seen": 74842112 }, { "epoch": 0.02, "learning_rate": 0.0004935907723169509, "loss": 3.9026, "theoretical_loss": 4.9713600850058395, "tokens_seen": 74907648 }, { "epoch": 0.02, "learning_rate": 0.00049358074222668, "loss": 3.5683, "theoretical_loss": 4.970732711610667, "tokens_seen": 74973184 }, { "epoch": 0.02, "learning_rate": 0.0004935707121364092, "loss": 3.6196, "theoretical_loss": 4.97010603977933, "tokens_seen": 75038720 }, { "epoch": 0.02, "learning_rate": 0.0004935606820461384, "loss": 3.8518, "theoretical_loss": 4.96948006811561, "tokens_seen": 75104256 }, { "epoch": 0.02, "learning_rate": 0.0004935506519558676, "loss": 3.8093, "theoretical_loss": 4.968854795227281, "tokens_seen": 75169792 }, { "epoch": 0.02, "learning_rate": 0.0004935406218655968, "loss": 3.7438, "theoretical_loss": 4.968230219726093, "tokens_seen": 75235328 }, { "epoch": 0.02, "learning_rate": 0.000493530591775326, "loss": 3.6534, "theoretical_loss": 4.967606340227765, "tokens_seen": 75300864 }, { "epoch": 0.02, "objective/train/docs_used": 54881, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.100285053253174, "objective/train/theoretical_loss": 4.966983155351962, "objective/train/tokens_used": 95826400, "theoretical_loss": 4.966983155351962, "tokens_seen": 75366400 }, { "epoch": 0.02, "learning_rate": 0.0004935205616850551, "loss": 3.7242, "theoretical_loss": 4.966983155351962, "tokens_seen": 75366400 }, { "epoch": 0.02, "learning_rate": 0.0004935105315947844, "loss": 3.7258, "theoretical_loss": 4.966360663722287, "tokens_seen": 75431936 }, { "epoch": 0.02, "learning_rate": 0.0004935005015045135, "loss": 3.5934, "theoretical_loss": 4.96573886396626, "tokens_seen": 75497472 }, { "epoch": 0.02, "learning_rate": 0.0004934904714142428, "loss": 3.4784, "theoretical_loss": 4.965117754715307, "tokens_seen": 75563008 }, { "epoch": 0.02, "learning_rate": 0.0004934804413239719, "loss": 3.4207, "theoretical_loss": 4.964497334604748, "tokens_seen": 75628544 }, { "epoch": 0.02, "learning_rate": 0.0004934704112337011, "loss": 3.8166, "theoretical_loss": 4.963877602273776, "tokens_seen": 75694080 }, { "epoch": 0.02, "learning_rate": 0.0004934603811434303, "loss": 3.8047, "theoretical_loss": 4.963258556365449, "tokens_seen": 75759616 }, { "epoch": 0.02, "learning_rate": 0.0004934503510531595, "loss": 3.7029, "theoretical_loss": 4.962640195526673, "tokens_seen": 75825152 }, { "epoch": 0.02, "learning_rate": 0.0004934403209628887, "loss": 3.7634, "theoretical_loss": 4.962022518408183, "tokens_seen": 75890688 }, { "epoch": 0.02, "learning_rate": 0.0004934302908726179, "loss": 3.5837, "theoretical_loss": 4.96140552366454, "tokens_seen": 75956224 }, { "epoch": 0.02, "learning_rate": 0.000493420260782347, "loss": 3.6947, "theoretical_loss": 4.9607892099541075, "tokens_seen": 76021760 }, { "epoch": 0.02, "learning_rate": 0.0004934102306920762, "loss": 3.6329, "theoretical_loss": 4.9601735759390415, "tokens_seen": 76087296 }, { "epoch": 0.02, "learning_rate": 0.0004934002006018054, "loss": 3.615, "theoretical_loss": 4.959558620285274, "tokens_seen": 76152832 }, { "epoch": 0.02, "learning_rate": 0.0004933901705115346, "loss": 3.7512, "theoretical_loss": 4.958944341662502, "tokens_seen": 76218368 }, { "epoch": 0.02, "learning_rate": 0.0004933801404212638, "loss": 3.6382, "theoretical_loss": 4.958330738744172, "tokens_seen": 76283904 }, { "epoch": 0.02, "learning_rate": 0.000493370110330993, "loss": 3.7997, "theoretical_loss": 4.957717810207466, "tokens_seen": 76349440 }, { "epoch": 0.02, "learning_rate": 0.0004933600802407221, "loss": 3.6714, "theoretical_loss": 4.957105554733289, "tokens_seen": 76414976 }, { "epoch": 0.02, "learning_rate": 0.0004933500501504513, "loss": 3.4102, "theoretical_loss": 4.956493971006253, "tokens_seen": 76480512 }, { "epoch": 0.02, "learning_rate": 0.0004933400200601805, "loss": 3.6513, "theoretical_loss": 4.955883057714669, "tokens_seen": 76546048 }, { "epoch": 0.02, "learning_rate": 0.0004933299899699098, "loss": 3.7063, "theoretical_loss": 4.955272813550524, "tokens_seen": 76611584 }, { "epoch": 0.02, "learning_rate": 0.0004933199598796389, "loss": 3.5682, "theoretical_loss": 4.954663237209477, "tokens_seen": 76677120 }, { "epoch": 0.02, "learning_rate": 0.0004933099297893682, "loss": 3.859, "theoretical_loss": 4.954054327390841, "tokens_seen": 76742656 }, { "epoch": 0.02, "learning_rate": 0.0004932998996990972, "loss": 3.6904, "theoretical_loss": 4.9534460827975675, "tokens_seen": 76808192 }, { "epoch": 0.02, "learning_rate": 0.0004932898696088265, "loss": 3.5948, "theoretical_loss": 4.952838502136241, "tokens_seen": 76873728 }, { "epoch": 0.02, "learning_rate": 0.0004932798395185557, "loss": 3.6899, "theoretical_loss": 4.952231584117056, "tokens_seen": 76939264 }, { "epoch": 0.02, "objective/train/docs_used": 55424, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.752882957458496, "objective/train/theoretical_loss": 4.951625327453812, "objective/train/tokens_used": 97464800, "theoretical_loss": 4.951625327453812, "tokens_seen": 77004800 }, { "epoch": 0.02, "learning_rate": 0.0004932698094282849, "loss": 3.6065, "theoretical_loss": 4.951625327453812, "tokens_seen": 77004800 }, { "epoch": 0.02, "learning_rate": 0.0004932597793380141, "loss": 3.69, "theoretical_loss": 4.951019730863894, "tokens_seen": 77070336 }, { "epoch": 0.02, "learning_rate": 0.0004932497492477432, "loss": 3.5959, "theoretical_loss": 4.950414793068266, "tokens_seen": 77135872 }, { "epoch": 0.02, "learning_rate": 0.0004932397191574724, "loss": 3.7588, "theoretical_loss": 4.94981051279145, "tokens_seen": 77201408 }, { "epoch": 0.02, "learning_rate": 0.0004932296890672016, "loss": 3.8397, "theoretical_loss": 4.94920688876152, "tokens_seen": 77266944 }, { "epoch": 0.02, "learning_rate": 0.0004932196589769308, "loss": 3.4345, "theoretical_loss": 4.948603919710088, "tokens_seen": 77332480 }, { "epoch": 0.02, "learning_rate": 0.00049320962888666, "loss": 3.6935, "theoretical_loss": 4.948001604372287, "tokens_seen": 77398016 }, { "epoch": 0.02, "learning_rate": 0.0004931995987963893, "loss": 3.6558, "theoretical_loss": 4.947399941486762, "tokens_seen": 77463552 }, { "epoch": 0.02, "learning_rate": 0.0004931895687061183, "loss": 4.0196, "theoretical_loss": 4.946798929795658, "tokens_seen": 77529088 }, { "epoch": 0.02, "learning_rate": 0.0004931795386158476, "loss": 3.5714, "theoretical_loss": 4.946198568044602, "tokens_seen": 77594624 }, { "epoch": 0.02, "learning_rate": 0.0004931695085255767, "loss": 3.4172, "theoretical_loss": 4.945598854982698, "tokens_seen": 77660160 }, { "epoch": 0.02, "learning_rate": 0.000493159478435306, "loss": 3.8399, "theoretical_loss": 4.944999789362508, "tokens_seen": 77725696 }, { "epoch": 0.02, "learning_rate": 0.0004931494483450352, "loss": 3.7901, "theoretical_loss": 4.944401369940043, "tokens_seen": 77791232 }, { "epoch": 0.02, "learning_rate": 0.0004931394182547643, "loss": 3.5237, "theoretical_loss": 4.9438035954747495, "tokens_seen": 77856768 }, { "epoch": 0.02, "learning_rate": 0.0004931293881644935, "loss": 3.4995, "theoretical_loss": 4.9432064647294975, "tokens_seen": 77922304 }, { "epoch": 0.02, "learning_rate": 0.0004931193580742227, "loss": 3.8164, "theoretical_loss": 4.942609976470566, "tokens_seen": 77987840 }, { "epoch": 0.02, "learning_rate": 0.0004931093279839519, "loss": 3.5615, "theoretical_loss": 4.942014129467637, "tokens_seen": 78053376 }, { "epoch": 0.02, "learning_rate": 0.0004930992978936811, "loss": 3.3958, "theoretical_loss": 4.941418922493774, "tokens_seen": 78118912 }, { "epoch": 0.02, "learning_rate": 0.0004930892678034102, "loss": 3.5015, "theoretical_loss": 4.940824354325419, "tokens_seen": 78184448 }, { "epoch": 0.02, "learning_rate": 0.0004930792377131394, "loss": 3.5466, "theoretical_loss": 4.940230423742372, "tokens_seen": 78249984 }, { "epoch": 0.02, "learning_rate": 0.0004930692076228686, "loss": 3.503, "theoretical_loss": 4.939637129527789, "tokens_seen": 78315520 }, { "epoch": 0.02, "learning_rate": 0.0004930591775325978, "loss": 3.6357, "theoretical_loss": 4.939044470468156, "tokens_seen": 78381056 }, { "epoch": 0.02, "learning_rate": 0.000493049147442327, "loss": 3.7812, "theoretical_loss": 4.938452445353294, "tokens_seen": 78446592 }, { "epoch": 0.02, "learning_rate": 0.0004930391173520562, "loss": 3.4835, "theoretical_loss": 4.937861052976332, "tokens_seen": 78512128 }, { "epoch": 0.02, "learning_rate": 0.0004930290872617853, "loss": 3.1521, "theoretical_loss": 4.937270292133704, "tokens_seen": 78577664 }, { "epoch": 0.02, "objective/train/docs_used": 56825, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.739461898803711, "objective/train/theoretical_loss": 4.9366801616251355, "objective/train/tokens_used": 99103200, "theoretical_loss": 4.9366801616251355, "tokens_seen": 78643200 }, { "epoch": 0.02, "learning_rate": 0.0004930190571715146, "loss": 3.8046, "theoretical_loss": 4.9366801616251355, "tokens_seen": 78643200 }, { "epoch": 0.02, "learning_rate": 0.0004930090270812437, "loss": 3.5368, "theoretical_loss": 4.93609066025363, "tokens_seen": 78708736 }, { "epoch": 0.02, "learning_rate": 0.000492998996990973, "loss": 3.878, "theoretical_loss": 4.935501786825457, "tokens_seen": 78774272 }, { "epoch": 0.02, "learning_rate": 0.0004929889669007021, "loss": 3.6171, "theoretical_loss": 4.934913540150143, "tokens_seen": 78839808 }, { "epoch": 0.02, "learning_rate": 0.0004929789368104313, "loss": 3.5845, "theoretical_loss": 4.934325919040461, "tokens_seen": 78905344 }, { "epoch": 0.02, "learning_rate": 0.0004929689067201605, "loss": 3.5543, "theoretical_loss": 4.933738922312413, "tokens_seen": 78970880 }, { "epoch": 0.02, "learning_rate": 0.0004929588766298897, "loss": 3.3024, "theoretical_loss": 4.933152548785222, "tokens_seen": 79036416 }, { "epoch": 0.02, "learning_rate": 0.0004929488465396189, "loss": 3.6027, "theoretical_loss": 4.932566797281324, "tokens_seen": 79101952 }, { "epoch": 0.02, "learning_rate": 0.0004929388164493481, "loss": 3.9121, "theoretical_loss": 4.931981666626351, "tokens_seen": 79167488 }, { "epoch": 0.02, "learning_rate": 0.0004929287863590773, "loss": 3.5938, "theoretical_loss": 4.931397155649121, "tokens_seen": 79233024 }, { "epoch": 0.02, "learning_rate": 0.0004929187562688064, "loss": 3.7354, "theoretical_loss": 4.930813263181631, "tokens_seen": 79298560 }, { "epoch": 0.02, "learning_rate": 0.0004929087261785356, "loss": 3.433, "theoretical_loss": 4.93022998805904, "tokens_seen": 79364096 }, { "epoch": 0.02, "learning_rate": 0.0004928986960882648, "loss": 3.6705, "theoretical_loss": 4.929647329119659, "tokens_seen": 79429632 }, { "epoch": 0.02, "learning_rate": 0.000492888665997994, "loss": 3.3607, "theoretical_loss": 4.9290652852049455, "tokens_seen": 79495168 }, { "epoch": 0.02, "learning_rate": 0.0004928786359077232, "loss": 3.5638, "theoretical_loss": 4.928483855159485, "tokens_seen": 79560704 }, { "epoch": 0.02, "learning_rate": 0.0004928686058174523, "loss": 3.8926, "theoretical_loss": 4.927903037830983, "tokens_seen": 79626240 }, { "epoch": 0.02, "learning_rate": 0.0004928585757271815, "loss": 3.8163, "theoretical_loss": 4.9273228320702565, "tokens_seen": 79691776 }, { "epoch": 0.02, "learning_rate": 0.0004928485456369107, "loss": 3.6785, "theoretical_loss": 4.926743236731218, "tokens_seen": 79757312 }, { "epoch": 0.02, "learning_rate": 0.00049283851554664, "loss": 3.5302, "theoretical_loss": 4.926164250670868, "tokens_seen": 79822848 }, { "epoch": 0.02, "learning_rate": 0.0004928284854563691, "loss": 3.4988, "theoretical_loss": 4.925585872749284, "tokens_seen": 79888384 }, { "epoch": 0.02, "learning_rate": 0.0004928184553660984, "loss": 3.6865, "theoretical_loss": 4.925008101829608, "tokens_seen": 79953920 }, { "epoch": 0.02, "learning_rate": 0.0004928084252758274, "loss": 3.5066, "theoretical_loss": 4.9244309367780374, "tokens_seen": 80019456 }, { "epoch": 0.02, "learning_rate": 0.0004927983951855567, "loss": 3.8955, "theoretical_loss": 4.923854376463816, "tokens_seen": 80084992 }, { "epoch": 0.02, "learning_rate": 0.0004927883650952859, "loss": 3.7358, "theoretical_loss": 4.923278419759217, "tokens_seen": 80150528 }, { "epoch": 0.02, "learning_rate": 0.0004927783350050151, "loss": 3.7513, "theoretical_loss": 4.92270306553954, "tokens_seen": 80216064 }, { "epoch": 0.02, "objective/train/docs_used": 57296, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.519014596939087, "objective/train/theoretical_loss": 4.922128312683096, "objective/train/tokens_used": 100741600, "theoretical_loss": 4.922128312683096, "tokens_seen": 80281600 }, { "epoch": 0.02, "learning_rate": 0.0004927683049147443, "loss": 3.5332, "theoretical_loss": 4.922128312683096, "tokens_seen": 80281600 }, { "epoch": 0.02, "learning_rate": 0.0004927582748244734, "loss": 3.5115, "theoretical_loss": 4.921554160071194, "tokens_seen": 80347136 }, { "epoch": 0.02, "learning_rate": 0.0004927482447342026, "loss": 3.4584, "theoretical_loss": 4.920980606588142, "tokens_seen": 80412672 }, { "epoch": 0.02, "learning_rate": 0.0004927382146439318, "loss": 3.6266, "theoretical_loss": 4.920407651121222, "tokens_seen": 80478208 }, { "epoch": 0.02, "learning_rate": 0.000492728184553661, "loss": 3.8078, "theoretical_loss": 4.919835292560689, "tokens_seen": 80543744 }, { "epoch": 0.02, "learning_rate": 0.0004927181544633902, "loss": 3.8442, "theoretical_loss": 4.919263529799759, "tokens_seen": 80609280 }, { "epoch": 0.02, "learning_rate": 0.0004927081243731193, "loss": 3.8532, "theoretical_loss": 4.918692361734598, "tokens_seen": 80674816 }, { "epoch": 0.02, "learning_rate": 0.0004926980942828485, "loss": 3.4396, "theoretical_loss": 4.91812178726431, "tokens_seen": 80740352 }, { "epoch": 0.02, "learning_rate": 0.0004926880641925777, "loss": 3.6955, "theoretical_loss": 4.917551805290929, "tokens_seen": 80805888 }, { "epoch": 0.02, "learning_rate": 0.0004926780341023069, "loss": 3.5437, "theoretical_loss": 4.916982414719408, "tokens_seen": 80871424 }, { "epoch": 0.02, "learning_rate": 0.0004926680040120361, "loss": 3.8737, "theoretical_loss": 4.9164136144576105, "tokens_seen": 80936960 }, { "epoch": 0.02, "learning_rate": 0.0004926579739217654, "loss": 3.464, "theoretical_loss": 4.915845403416299, "tokens_seen": 81002496 }, { "epoch": 0.02, "learning_rate": 0.0004926479438314944, "loss": 3.6247, "theoretical_loss": 4.915277780509124, "tokens_seen": 81068032 }, { "epoch": 0.02, "learning_rate": 0.0004926379137412237, "loss": 3.4444, "theoretical_loss": 4.914710744652614, "tokens_seen": 81133568 }, { "epoch": 0.02, "learning_rate": 0.0004926278836509528, "loss": 3.5921, "theoretical_loss": 4.914144294766169, "tokens_seen": 81199104 }, { "epoch": 0.02, "learning_rate": 0.0004926178535606821, "loss": 3.5012, "theoretical_loss": 4.913578429772047, "tokens_seen": 81264640 }, { "epoch": 0.02, "learning_rate": 0.0004926078234704113, "loss": 3.3114, "theoretical_loss": 4.913013148595355, "tokens_seen": 81330176 }, { "epoch": 0.02, "learning_rate": 0.0004925977933801404, "loss": 3.7158, "theoretical_loss": 4.912448450164041, "tokens_seen": 81395712 }, { "epoch": 0.02, "learning_rate": 0.0004925877632898696, "loss": 3.7015, "theoretical_loss": 4.91188433340888, "tokens_seen": 81461248 }, { "epoch": 0.02, "learning_rate": 0.0004925777331995988, "loss": 3.9955, "theoretical_loss": 4.911320797263471, "tokens_seen": 81526784 }, { "epoch": 0.02, "learning_rate": 0.000492567703109328, "loss": 3.8269, "theoretical_loss": 4.910757840664219, "tokens_seen": 81592320 }, { "epoch": 0.02, "learning_rate": 0.0004925576730190572, "loss": 3.5523, "theoretical_loss": 4.910195462550334, "tokens_seen": 81657856 }, { "epoch": 0.02, "learning_rate": 0.0004925476429287864, "loss": 3.552, "theoretical_loss": 4.909633661863811, "tokens_seen": 81723392 }, { "epoch": 0.02, "learning_rate": 0.0004925376128385155, "loss": 3.8094, "theoretical_loss": 4.909072437549434, "tokens_seen": 81788928 }, { "epoch": 0.02, "learning_rate": 0.0004925275827482447, "loss": 3.7451, "theoretical_loss": 4.908511788554753, "tokens_seen": 81854464 }, { "epoch": 0.02, "objective/train/docs_used": 58605, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.980154275894165, "objective/train/theoretical_loss": 4.907951713830082, "objective/train/tokens_used": 102380000, "theoretical_loss": 4.907951713830082, "tokens_seen": 81920000 }, { "epoch": 0.02, "learning_rate": 0.0004925175526579739, "loss": 3.833, "theoretical_loss": 4.907951713830082, "tokens_seen": 81920000 }, { "epoch": 0.02, "learning_rate": 0.0004925075225677031, "loss": 3.5955, "theoretical_loss": 4.907392212328489, "tokens_seen": 81985536 }, { "epoch": 0.02, "learning_rate": 0.0004924974924774323, "loss": 3.3892, "theoretical_loss": 4.906833283005785, "tokens_seen": 82051072 }, { "epoch": 0.02, "learning_rate": 0.0004924874623871615, "loss": 3.7307, "theoretical_loss": 4.906274924820515, "tokens_seen": 82116608 }, { "epoch": 0.02, "learning_rate": 0.0004924774322968907, "loss": 3.7447, "theoretical_loss": 4.90571713673395, "tokens_seen": 82182144 }, { "epoch": 0.02, "learning_rate": 0.0004924674022066199, "loss": 3.5136, "theoretical_loss": 4.905159917710073, "tokens_seen": 82247680 }, { "epoch": 0.02, "learning_rate": 0.0004924573721163491, "loss": 3.6628, "theoretical_loss": 4.904603266715578, "tokens_seen": 82313216 }, { "epoch": 0.02, "learning_rate": 0.0004924473420260783, "loss": 3.7499, "theoretical_loss": 4.904047182719854, "tokens_seen": 82378752 }, { "epoch": 0.02, "learning_rate": 0.0004924373119358075, "loss": 3.4528, "theoretical_loss": 4.903491664694977, "tokens_seen": 82444288 }, { "epoch": 0.03, "learning_rate": 0.0004924272818455366, "loss": 3.614, "theoretical_loss": 4.902936711615702, "tokens_seen": 82509824 }, { "epoch": 0.03, "learning_rate": 0.0004924172517552658, "loss": 3.7331, "theoretical_loss": 4.902382322459456, "tokens_seen": 82575360 }, { "epoch": 0.03, "learning_rate": 0.000492407221664995, "loss": 3.5698, "theoretical_loss": 4.901828496206322, "tokens_seen": 82640896 }, { "epoch": 0.03, "learning_rate": 0.0004923971915747242, "loss": 3.6838, "theoretical_loss": 4.90127523183904, "tokens_seen": 82706432 }, { "epoch": 0.03, "learning_rate": 0.0004923871614844534, "loss": 3.7826, "theoretical_loss": 4.900722528342988, "tokens_seen": 82771968 }, { "epoch": 0.03, "learning_rate": 0.0004923771313941825, "loss": 3.6392, "theoretical_loss": 4.900170384706181, "tokens_seen": 82837504 }, { "epoch": 0.03, "learning_rate": 0.0004923671013039117, "loss": 3.491, "theoretical_loss": 4.899618799919256, "tokens_seen": 82903040 }, { "epoch": 0.03, "learning_rate": 0.0004923570712136409, "loss": 3.4612, "theoretical_loss": 4.899067772975469, "tokens_seen": 82968576 }, { "epoch": 0.03, "learning_rate": 0.0004923470411233702, "loss": 3.687, "theoretical_loss": 4.898517302870679, "tokens_seen": 83034112 }, { "epoch": 0.03, "learning_rate": 0.0004923370110330993, "loss": 3.7585, "theoretical_loss": 4.897967388603346, "tokens_seen": 83099648 }, { "epoch": 0.03, "learning_rate": 0.0004923269809428286, "loss": 3.5586, "theoretical_loss": 4.897418029174519, "tokens_seen": 83165184 }, { "epoch": 0.03, "learning_rate": 0.0004923169508525576, "loss": 3.9548, "theoretical_loss": 4.896869223587828, "tokens_seen": 83230720 }, { "epoch": 0.03, "learning_rate": 0.0004923069207622869, "loss": 3.8464, "theoretical_loss": 4.896320970849472, "tokens_seen": 83296256 }, { "epoch": 0.03, "learning_rate": 0.0004922968906720161, "loss": 3.5297, "theoretical_loss": 4.895773269968219, "tokens_seen": 83361792 }, { "epoch": 0.03, "learning_rate": 0.0004922868605817453, "loss": 3.7033, "theoretical_loss": 4.895226119955386, "tokens_seen": 83427328 }, { "epoch": 0.03, "learning_rate": 0.0004922768304914745, "loss": 3.9995, "theoretical_loss": 4.894679519824841, "tokens_seen": 83492864 }, { "epoch": 0.03, "objective/train/docs_used": 59310, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.926790952682495, "objective/train/theoretical_loss": 4.894133468592984, "objective/train/tokens_used": 104018400, "theoretical_loss": 4.894133468592984, "tokens_seen": 83558400 }, { "epoch": 0.03, "learning_rate": 0.0004922668004012036, "loss": 3.6992, "theoretical_loss": 4.894133468592984, "tokens_seen": 83558400 }, { "epoch": 0.03, "learning_rate": 0.0004922567703109328, "loss": 3.6724, "theoretical_loss": 4.8935879652787495, "tokens_seen": 83623936 }, { "epoch": 0.03, "learning_rate": 0.000492246740220662, "loss": 3.7634, "theoretical_loss": 4.893043008903591, "tokens_seen": 83689472 }, { "epoch": 0.03, "learning_rate": 0.0004922367101303912, "loss": 3.701, "theoretical_loss": 4.892498598491473, "tokens_seen": 83755008 }, { "epoch": 0.03, "learning_rate": 0.0004922266800401204, "loss": 3.6549, "theoretical_loss": 4.891954733068863, "tokens_seen": 83820544 }, { "epoch": 0.03, "learning_rate": 0.0004922166499498495, "loss": 3.6165, "theoretical_loss": 4.891411411664727, "tokens_seen": 83886080 }, { "epoch": 0.03, "learning_rate": 0.0004922066198595787, "loss": 3.682, "theoretical_loss": 4.890868633310515, "tokens_seen": 83951616 }, { "epoch": 0.03, "learning_rate": 0.0004921965897693079, "loss": 3.4424, "theoretical_loss": 4.890326397040158, "tokens_seen": 84017152 }, { "epoch": 0.03, "learning_rate": 0.0004921865596790371, "loss": 3.6114, "theoretical_loss": 4.889784701890056, "tokens_seen": 84082688 }, { "epoch": 0.03, "learning_rate": 0.0004921765295887663, "loss": 3.5193, "theoretical_loss": 4.8892435468990705, "tokens_seen": 84148224 }, { "epoch": 0.03, "learning_rate": 0.0004921664994984956, "loss": 3.2842, "theoretical_loss": 4.88870293110852, "tokens_seen": 84213760 }, { "epoch": 0.03, "learning_rate": 0.0004921564694082246, "loss": 3.7096, "theoretical_loss": 4.888162853562166, "tokens_seen": 84279296 }, { "epoch": 0.03, "learning_rate": 0.0004921464393179539, "loss": 3.5071, "theoretical_loss": 4.88762331330621, "tokens_seen": 84344832 }, { "epoch": 0.03, "learning_rate": 0.000492136409227683, "loss": 3.4462, "theoretical_loss": 4.88708430938928, "tokens_seen": 84410368 }, { "epoch": 0.03, "learning_rate": 0.0004921263791374123, "loss": 3.7536, "theoretical_loss": 4.8865458408624285, "tokens_seen": 84475904 }, { "epoch": 0.03, "learning_rate": 0.0004921163490471415, "loss": 3.493, "theoretical_loss": 4.8860079067791204, "tokens_seen": 84541440 }, { "epoch": 0.03, "learning_rate": 0.0004921063189568706, "loss": 3.7587, "theoretical_loss": 4.885470506195227, "tokens_seen": 84606976 }, { "epoch": 0.03, "learning_rate": 0.0004920962888665998, "loss": 3.6445, "theoretical_loss": 4.884933638169014, "tokens_seen": 84672512 }, { "epoch": 0.03, "learning_rate": 0.000492086258776329, "loss": 3.5871, "theoretical_loss": 4.88439730176114, "tokens_seen": 84738048 }, { "epoch": 0.03, "learning_rate": 0.0004920762286860582, "loss": 3.772, "theoretical_loss": 4.883861496034644, "tokens_seen": 84803584 }, { "epoch": 0.03, "learning_rate": 0.0004920661985957874, "loss": 3.4383, "theoretical_loss": 4.88332622005494, "tokens_seen": 84869120 }, { "epoch": 0.03, "learning_rate": 0.0004920561685055166, "loss": 3.4985, "theoretical_loss": 4.8827914728898065, "tokens_seen": 84934656 }, { "epoch": 0.03, "learning_rate": 0.0004920461384152457, "loss": 3.5046, "theoretical_loss": 4.88225725360938, "tokens_seen": 85000192 }, { "epoch": 0.03, "learning_rate": 0.0004920361083249749, "loss": 3.682, "theoretical_loss": 4.881723561286149, "tokens_seen": 85065728 }, { "epoch": 0.03, "learning_rate": 0.0004920260782347041, "loss": 3.422, "theoretical_loss": 4.881190394994943, "tokens_seen": 85131264 }, { "epoch": 0.03, "objective/train/docs_used": 60492, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.639910936355591, "objective/train/theoretical_loss": 4.880657753812926, "objective/train/tokens_used": 105656800, "theoretical_loss": 4.880657753812926, "tokens_seen": 85196800 }, { "epoch": 0.03, "learning_rate": 0.0004920160481444333, "loss": 3.7149, "theoretical_loss": 4.880657753812926, "tokens_seen": 85196800 }, { "epoch": 0.03, "learning_rate": 0.0004920060180541625, "loss": 3.5313, "theoretical_loss": 4.880125636819594, "tokens_seen": 85262336 }, { "epoch": 0.03, "learning_rate": 0.0004919959879638916, "loss": 3.6137, "theoretical_loss": 4.879594043096755, "tokens_seen": 85327872 }, { "epoch": 0.03, "learning_rate": 0.0004919859578736209, "loss": 3.4991, "theoretical_loss": 4.879062971728534, "tokens_seen": 85393408 }, { "epoch": 0.03, "learning_rate": 0.00049197592778335, "loss": 3.6767, "theoretical_loss": 4.87853242180136, "tokens_seen": 85458944 }, { "epoch": 0.03, "learning_rate": 0.0004919658976930793, "loss": 3.7686, "theoretical_loss": 4.878002392403959, "tokens_seen": 85524480 }, { "epoch": 0.03, "learning_rate": 0.0004919558676028084, "loss": 3.5113, "theoretical_loss": 4.877472882627343, "tokens_seen": 85590016 }, { "epoch": 0.03, "learning_rate": 0.0004919458375125377, "loss": 3.7634, "theoretical_loss": 4.8769438915648085, "tokens_seen": 85655552 }, { "epoch": 0.03, "learning_rate": 0.0004919358074222668, "loss": 3.4673, "theoretical_loss": 4.876415418311928, "tokens_seen": 85721088 }, { "epoch": 0.03, "learning_rate": 0.000491925777331996, "loss": 3.562, "theoretical_loss": 4.875887461966537, "tokens_seen": 85786624 }, { "epoch": 0.03, "learning_rate": 0.0004919157472417252, "loss": 3.7082, "theoretical_loss": 4.875360021628733, "tokens_seen": 85852160 }, { "epoch": 0.03, "learning_rate": 0.0004919057171514544, "loss": 3.558, "theoretical_loss": 4.874833096400865, "tokens_seen": 85917696 }, { "epoch": 0.03, "learning_rate": 0.0004918956870611836, "loss": 3.259, "theoretical_loss": 4.874306685387525, "tokens_seen": 85983232 }, { "epoch": 0.03, "learning_rate": 0.0004918856569709127, "loss": 3.6862, "theoretical_loss": 4.873780787695547, "tokens_seen": 86048768 }, { "epoch": 0.03, "learning_rate": 0.0004918756268806419, "loss": 3.546, "theoretical_loss": 4.87325540243399, "tokens_seen": 86114304 }, { "epoch": 0.03, "learning_rate": 0.0004918655967903711, "loss": 3.5487, "theoretical_loss": 4.872730528714139, "tokens_seen": 86179840 }, { "epoch": 0.03, "learning_rate": 0.0004918555667001003, "loss": 3.57, "theoretical_loss": 4.872206165649493, "tokens_seen": 86245376 }, { "epoch": 0.03, "learning_rate": 0.0004918455366098295, "loss": 3.6859, "theoretical_loss": 4.871682312355761, "tokens_seen": 86310912 }, { "epoch": 0.03, "learning_rate": 0.0004918355065195586, "loss": 3.588, "theoretical_loss": 4.871158967950852, "tokens_seen": 86376448 }, { "epoch": 0.03, "learning_rate": 0.0004918254764292878, "loss": 3.6822, "theoretical_loss": 4.870636131554869, "tokens_seen": 86441984 }, { "epoch": 0.03, "learning_rate": 0.000491815446339017, "loss": 3.4897, "theoretical_loss": 4.8701138022901045, "tokens_seen": 86507520 }, { "epoch": 0.03, "learning_rate": 0.0004918054162487463, "loss": 3.5301, "theoretical_loss": 4.869591979281028, "tokens_seen": 86573056 }, { "epoch": 0.03, "learning_rate": 0.0004917953861584754, "loss": 3.3777, "theoretical_loss": 4.8690706616542805, "tokens_seen": 86638592 }, { "epoch": 0.03, "learning_rate": 0.0004917853560682047, "loss": 3.3358, "theoretical_loss": 4.868549848538675, "tokens_seen": 86704128 }, { "epoch": 0.03, "learning_rate": 0.0004917753259779337, "loss": 3.3347, "theoretical_loss": 4.868029539065176, "tokens_seen": 86769664 }, { "epoch": 0.03, "objective/train/docs_used": 61005, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4489526748657227, "objective/train/theoretical_loss": 4.867509732366907, "objective/train/tokens_used": 107295200, "theoretical_loss": 4.867509732366907, "tokens_seen": 86835200 }, { "epoch": 0.03, "learning_rate": 0.000491765295887663, "loss": 3.6654, "theoretical_loss": 4.867509732366907, "tokens_seen": 86835200 }, { "epoch": 0.03, "learning_rate": 0.0004917552657973922, "loss": 3.8192, "theoretical_loss": 4.866990427579129, "tokens_seen": 86900736 }, { "epoch": 0.03, "learning_rate": 0.0004917452357071214, "loss": 3.5657, "theoretical_loss": 4.866471623839248, "tokens_seen": 86966272 }, { "epoch": 0.03, "learning_rate": 0.0004917352056168506, "loss": 3.348, "theoretical_loss": 4.8659533202867955, "tokens_seen": 87031808 }, { "epoch": 0.03, "learning_rate": 0.0004917251755265797, "loss": 3.4467, "theoretical_loss": 4.86543551606343, "tokens_seen": 87097344 }, { "epoch": 0.03, "learning_rate": 0.0004917151454363089, "loss": 3.5397, "theoretical_loss": 4.864918210312927, "tokens_seen": 87162880 }, { "epoch": 0.03, "learning_rate": 0.0004917051153460381, "loss": 3.5919, "theoretical_loss": 4.864401402181173, "tokens_seen": 87228416 }, { "epoch": 0.03, "learning_rate": 0.0004916950852557673, "loss": 3.5553, "theoretical_loss": 4.863885090816158, "tokens_seen": 87293952 }, { "epoch": 0.03, "learning_rate": 0.0004916850551654965, "loss": 3.6125, "theoretical_loss": 4.863369275367968, "tokens_seen": 87359488 }, { "epoch": 0.03, "learning_rate": 0.0004916750250752258, "loss": 3.4756, "theoretical_loss": 4.862853954988781, "tokens_seen": 87425024 }, { "epoch": 0.03, "learning_rate": 0.0004916649949849548, "loss": 3.5864, "theoretical_loss": 4.862339128832857, "tokens_seen": 87490560 }, { "epoch": 0.03, "learning_rate": 0.0004916549648946841, "loss": 3.6655, "theoretical_loss": 4.861824796056533, "tokens_seen": 87556096 }, { "epoch": 0.03, "learning_rate": 0.0004916449348044132, "loss": 3.5427, "theoretical_loss": 4.861310955818219, "tokens_seen": 87621632 }, { "epoch": 0.03, "learning_rate": 0.0004916349047141425, "loss": 3.5534, "theoretical_loss": 4.860797607278385, "tokens_seen": 87687168 }, { "epoch": 0.03, "learning_rate": 0.0004916248746238717, "loss": 3.7963, "theoretical_loss": 4.86028474959956, "tokens_seen": 87752704 }, { "epoch": 0.03, "learning_rate": 0.0004916148445336008, "loss": 3.6729, "theoretical_loss": 4.859772381946323, "tokens_seen": 87818240 }, { "epoch": 0.03, "learning_rate": 0.00049160481444333, "loss": 3.6349, "theoretical_loss": 4.859260503485298, "tokens_seen": 87883776 }, { "epoch": 0.03, "learning_rate": 0.0004915947843530592, "loss": 3.7414, "theoretical_loss": 4.858749113385144, "tokens_seen": 87949312 }, { "epoch": 0.03, "learning_rate": 0.0004915847542627884, "loss": 3.2873, "theoretical_loss": 4.858238210816554, "tokens_seen": 88014848 }, { "epoch": 0.03, "learning_rate": 0.0004915747241725176, "loss": 3.5819, "theoretical_loss": 4.8577277949522415, "tokens_seen": 88080384 }, { "epoch": 0.03, "learning_rate": 0.0004915646940822468, "loss": 3.6767, "theoretical_loss": 4.857217864966943, "tokens_seen": 88145920 }, { "epoch": 0.03, "learning_rate": 0.0004915546639919759, "loss": 3.3539, "theoretical_loss": 4.856708420037402, "tokens_seen": 88211456 }, { "epoch": 0.03, "learning_rate": 0.0004915446339017051, "loss": 3.8479, "theoretical_loss": 4.8561994593423705, "tokens_seen": 88276992 }, { "epoch": 0.03, "learning_rate": 0.0004915346038114343, "loss": 3.5546, "theoretical_loss": 4.8556909820625975, "tokens_seen": 88342528 }, { "epoch": 0.03, "learning_rate": 0.0004915245737211635, "loss": 3.8083, "theoretical_loss": 4.855182987380823, "tokens_seen": 88408064 }, { "epoch": 0.03, "objective/train/docs_used": 61597, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.9370856285095215, "objective/train/theoretical_loss": 4.854675474481779, "objective/train/tokens_used": 108933600, "theoretical_loss": 4.854675474481779, "tokens_seen": 88473600 }, { "epoch": 0.03, "learning_rate": 0.0004915145436308927, "loss": 3.6345, "theoretical_loss": 4.854675474481779, "tokens_seen": 88473600 }, { "epoch": 0.03, "learning_rate": 0.0004915045135406218, "loss": 3.4992, "theoretical_loss": 4.8541684425521705, "tokens_seen": 88539136 }, { "epoch": 0.03, "learning_rate": 0.0004914944834503511, "loss": 3.4698, "theoretical_loss": 4.85366189078068, "tokens_seen": 88604672 }, { "epoch": 0.03, "learning_rate": 0.0004914844533600802, "loss": 3.5404, "theoretical_loss": 4.853155818357957, "tokens_seen": 88670208 }, { "epoch": 0.03, "learning_rate": 0.0004914744232698095, "loss": 3.5519, "theoretical_loss": 4.852650224476609, "tokens_seen": 88735744 }, { "epoch": 0.03, "learning_rate": 0.0004914643931795386, "loss": 3.2668, "theoretical_loss": 4.852145108331205, "tokens_seen": 88801280 }, { "epoch": 0.03, "learning_rate": 0.0004914543630892679, "loss": 3.7668, "theoretical_loss": 4.851640469118255, "tokens_seen": 88866816 }, { "epoch": 0.03, "learning_rate": 0.000491444332998997, "loss": 3.6371, "theoretical_loss": 4.851136306036219, "tokens_seen": 88932352 }, { "epoch": 0.03, "learning_rate": 0.0004914343029087262, "loss": 3.5714, "theoretical_loss": 4.850632618285486, "tokens_seen": 88997888 }, { "epoch": 0.03, "learning_rate": 0.0004914242728184554, "loss": 3.4973, "theoretical_loss": 4.850129405068383, "tokens_seen": 89063424 }, { "epoch": 0.03, "learning_rate": 0.0004914142427281846, "loss": 3.6218, "theoretical_loss": 4.849626665589156, "tokens_seen": 89128960 }, { "epoch": 0.03, "learning_rate": 0.0004914042126379138, "loss": 3.3802, "theoretical_loss": 4.849124399053969, "tokens_seen": 89194496 }, { "epoch": 0.03, "learning_rate": 0.0004913941825476429, "loss": 3.456, "theoretical_loss": 4.8486226046709024, "tokens_seen": 89260032 }, { "epoch": 0.03, "learning_rate": 0.0004913841524573721, "loss": 3.6214, "theoretical_loss": 4.8481212816499415, "tokens_seen": 89325568 }, { "epoch": 0.03, "learning_rate": 0.0004913741223671013, "loss": 3.7055, "theoretical_loss": 4.847620429202967, "tokens_seen": 89391104 }, { "epoch": 0.03, "learning_rate": 0.0004913640922768305, "loss": 3.6888, "theoretical_loss": 4.847120046543763, "tokens_seen": 89456640 }, { "epoch": 0.03, "learning_rate": 0.0004913540621865597, "loss": 3.803, "theoretical_loss": 4.846620132887992, "tokens_seen": 89522176 }, { "epoch": 0.03, "learning_rate": 0.0004913440320962888, "loss": 3.392, "theoretical_loss": 4.8461206874532055, "tokens_seen": 89587712 }, { "epoch": 0.03, "learning_rate": 0.000491334002006018, "loss": 3.5634, "theoretical_loss": 4.845621709458831, "tokens_seen": 89653248 }, { "epoch": 0.03, "learning_rate": 0.0004913239719157472, "loss": 3.6485, "theoretical_loss": 4.845123198126162, "tokens_seen": 89718784 }, { "epoch": 0.03, "learning_rate": 0.0004913139418254765, "loss": 3.6994, "theoretical_loss": 4.844625152678364, "tokens_seen": 89784320 }, { "epoch": 0.03, "learning_rate": 0.0004913039117352056, "loss": 3.6592, "theoretical_loss": 4.844127572340455, "tokens_seen": 89849856 }, { "epoch": 0.03, "learning_rate": 0.0004912938816449349, "loss": 3.6714, "theoretical_loss": 4.84363045633931, "tokens_seen": 89915392 }, { "epoch": 0.03, "learning_rate": 0.0004912838515546639, "loss": 3.3055, "theoretical_loss": 4.843133803903651, "tokens_seen": 89980928 }, { "epoch": 0.03, "learning_rate": 0.0004912738214643932, "loss": 3.6851, "theoretical_loss": 4.84263761426404, "tokens_seen": 90046464 }, { "epoch": 0.03, "objective/train/docs_used": 63098, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.415714979171753, "objective/train/theoretical_loss": 4.842141886652876, "objective/train/tokens_used": 110572000, "theoretical_loss": 4.842141886652876, "tokens_seen": 90112000 }, { "epoch": 0.03, "learning_rate": 0.0004912637913741224, "loss": 3.496, "theoretical_loss": 4.842141886652876, "tokens_seen": 90112000 }, { "epoch": 0.03, "learning_rate": 0.0004912537612838516, "loss": 3.7029, "theoretical_loss": 4.841646620304388, "tokens_seen": 90177536 }, { "epoch": 0.03, "learning_rate": 0.0004912437311935808, "loss": 3.4109, "theoretical_loss": 4.841151814454632, "tokens_seen": 90243072 }, { "epoch": 0.03, "learning_rate": 0.00049123370110331, "loss": 3.5139, "theoretical_loss": 4.840657468341476, "tokens_seen": 90308608 }, { "epoch": 0.03, "learning_rate": 0.0004912236710130391, "loss": 3.7375, "theoretical_loss": 4.84016358120461, "tokens_seen": 90374144 }, { "epoch": 0.03, "learning_rate": 0.0004912136409227683, "loss": 3.4378, "theoretical_loss": 4.839670152285526, "tokens_seen": 90439680 }, { "epoch": 0.03, "learning_rate": 0.0004912036108324975, "loss": 3.4727, "theoretical_loss": 4.8391771808275195, "tokens_seen": 90505216 }, { "epoch": 0.03, "learning_rate": 0.0004911935807422267, "loss": 3.4128, "theoretical_loss": 4.838684666075682, "tokens_seen": 90570752 }, { "epoch": 0.03, "learning_rate": 0.0004911835506519559, "loss": 3.5232, "theoretical_loss": 4.838192607276896, "tokens_seen": 90636288 }, { "epoch": 0.03, "learning_rate": 0.000491173520561685, "loss": 3.5274, "theoretical_loss": 4.837701003679829, "tokens_seen": 90701824 }, { "epoch": 0.03, "learning_rate": 0.0004911634904714142, "loss": 3.573, "theoretical_loss": 4.8372098545349305, "tokens_seen": 90767360 }, { "epoch": 0.03, "learning_rate": 0.0004911534603811434, "loss": 3.4979, "theoretical_loss": 4.836719159094422, "tokens_seen": 90832896 }, { "epoch": 0.03, "learning_rate": 0.0004911434302908726, "loss": 3.7194, "theoretical_loss": 4.836228916612292, "tokens_seen": 90898432 }, { "epoch": 0.03, "learning_rate": 0.0004911334002006019, "loss": 3.71, "theoretical_loss": 4.835739126344298, "tokens_seen": 90963968 }, { "epoch": 0.03, "learning_rate": 0.0004911233701103309, "loss": 3.4962, "theoretical_loss": 4.8352497875479505, "tokens_seen": 91029504 }, { "epoch": 0.03, "learning_rate": 0.0004911133400200602, "loss": 3.4739, "theoretical_loss": 4.834760899482514, "tokens_seen": 91095040 }, { "epoch": 0.03, "learning_rate": 0.0004911033099297893, "loss": 3.452, "theoretical_loss": 4.834272461409001, "tokens_seen": 91160576 }, { "epoch": 0.03, "learning_rate": 0.0004910932798395186, "loss": 3.8125, "theoretical_loss": 4.833784472590165, "tokens_seen": 91226112 }, { "epoch": 0.03, "learning_rate": 0.0004910832497492478, "loss": 3.3993, "theoretical_loss": 4.833296932290495, "tokens_seen": 91291648 }, { "epoch": 0.03, "learning_rate": 0.000491073219658977, "loss": 3.5064, "theoretical_loss": 4.832809839776213, "tokens_seen": 91357184 }, { "epoch": 0.03, "learning_rate": 0.0004910631895687061, "loss": 3.4702, "theoretical_loss": 4.832323194315265, "tokens_seen": 91422720 }, { "epoch": 0.03, "learning_rate": 0.0004910531594784353, "loss": 3.576, "theoretical_loss": 4.831836995177319, "tokens_seen": 91488256 }, { "epoch": 0.03, "learning_rate": 0.0004910431293881645, "loss": 3.6835, "theoretical_loss": 4.831351241633756, "tokens_seen": 91553792 }, { "epoch": 0.03, "learning_rate": 0.0004910330992978937, "loss": 3.6733, "theoretical_loss": 4.8308659329576695, "tokens_seen": 91619328 }, { "epoch": 0.03, "learning_rate": 0.0004910230692076229, "loss": 3.7039, "theoretical_loss": 4.830381068423856, "tokens_seen": 91684864 }, { "epoch": 0.03, "objective/train/docs_used": 63840, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2570016384124756, "objective/train/theoretical_loss": 4.8298966473088125, "objective/train/tokens_used": 112210400, "theoretical_loss": 4.8298966473088125, "tokens_seen": 91750400 }, { "epoch": 0.03, "learning_rate": 0.000491013039117352, "loss": 3.5049, "theoretical_loss": 4.8298966473088125, "tokens_seen": 91750400 }, { "epoch": 0.03, "learning_rate": 0.0004910030090270812, "loss": 3.7075, "theoretical_loss": 4.829412668890729, "tokens_seen": 91815936 }, { "epoch": 0.03, "learning_rate": 0.0004909929789368104, "loss": 3.3632, "theoretical_loss": 4.8289291324494865, "tokens_seen": 91881472 }, { "epoch": 0.03, "learning_rate": 0.0004909829488465397, "loss": 3.4331, "theoretical_loss": 4.828446037266647, "tokens_seen": 91947008 }, { "epoch": 0.03, "learning_rate": 0.0004909729187562688, "loss": 3.5859, "theoretical_loss": 4.827963382625454, "tokens_seen": 92012544 }, { "epoch": 0.03, "learning_rate": 0.0004909628886659981, "loss": 3.4441, "theoretical_loss": 4.827481167810825, "tokens_seen": 92078080 }, { "epoch": 0.03, "learning_rate": 0.0004909528585757272, "loss": 3.6517, "theoretical_loss": 4.826999392109344, "tokens_seen": 92143616 }, { "epoch": 0.03, "learning_rate": 0.0004909428284854564, "loss": 3.4679, "theoretical_loss": 4.826518054809259, "tokens_seen": 92209152 }, { "epoch": 0.03, "learning_rate": 0.0004909327983951856, "loss": 3.6007, "theoretical_loss": 4.826037155200478, "tokens_seen": 92274688 }, { "epoch": 0.03, "learning_rate": 0.0004909227683049148, "loss": 3.4858, "theoretical_loss": 4.825556692574562, "tokens_seen": 92340224 }, { "epoch": 0.03, "learning_rate": 0.000490912738214644, "loss": 3.5593, "theoretical_loss": 4.825076666224717, "tokens_seen": 92405760 }, { "epoch": 0.03, "learning_rate": 0.0004909027081243731, "loss": 3.5697, "theoretical_loss": 4.824597075445799, "tokens_seen": 92471296 }, { "epoch": 0.03, "learning_rate": 0.0004908926780341023, "loss": 3.5531, "theoretical_loss": 4.824117919534297, "tokens_seen": 92536832 }, { "epoch": 0.03, "learning_rate": 0.0004908826479438315, "loss": 3.5856, "theoretical_loss": 4.823639197788334, "tokens_seen": 92602368 }, { "epoch": 0.03, "learning_rate": 0.0004908726178535607, "loss": 3.6203, "theoretical_loss": 4.823160909507665, "tokens_seen": 92667904 }, { "epoch": 0.03, "learning_rate": 0.0004908625877632899, "loss": 3.5859, "theoretical_loss": 4.822683053993664, "tokens_seen": 92733440 }, { "epoch": 0.03, "learning_rate": 0.000490852557673019, "loss": 3.4464, "theoretical_loss": 4.822205630549329, "tokens_seen": 92798976 }, { "epoch": 0.03, "learning_rate": 0.0004908425275827482, "loss": 3.1732, "theoretical_loss": 4.821728638479267, "tokens_seen": 92864512 }, { "epoch": 0.03, "learning_rate": 0.0004908324974924774, "loss": 3.6278, "theoretical_loss": 4.821252077089696, "tokens_seen": 92930048 }, { "epoch": 0.03, "learning_rate": 0.0004908224674022067, "loss": 3.6351, "theoretical_loss": 4.820775945688437, "tokens_seen": 92995584 }, { "epoch": 0.03, "learning_rate": 0.0004908124373119358, "loss": 3.4334, "theoretical_loss": 4.820300243584913, "tokens_seen": 93061120 }, { "epoch": 0.03, "learning_rate": 0.0004908024072216651, "loss": 3.4374, "theoretical_loss": 4.819824970090138, "tokens_seen": 93126656 }, { "epoch": 0.03, "learning_rate": 0.0004907923771313941, "loss": 3.8253, "theoretical_loss": 4.819350124516717, "tokens_seen": 93192192 }, { "epoch": 0.03, "learning_rate": 0.0004907823470411234, "loss": 3.5773, "theoretical_loss": 4.818875706178841, "tokens_seen": 93257728 }, { "epoch": 0.03, "learning_rate": 0.0004907723169508526, "loss": 3.7615, "theoretical_loss": 4.818401714392279, "tokens_seen": 93323264 }, { "epoch": 0.03, "objective/train/docs_used": 65224, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.483311653137207, "objective/train/theoretical_loss": 4.817928148474378, "objective/train/tokens_used": 113848800, "theoretical_loss": 4.817928148474378, "tokens_seen": 93388800 }, { "epoch": 0.03, "learning_rate": 0.0004907622868605818, "loss": 3.571, "theoretical_loss": 4.817928148474378, "tokens_seen": 93388800 }, { "epoch": 0.03, "learning_rate": 0.000490752256770311, "loss": 3.5792, "theoretical_loss": 4.817455007744052, "tokens_seen": 93454336 }, { "epoch": 0.03, "learning_rate": 0.0004907422266800401, "loss": 3.4639, "theoretical_loss": 4.816982291521785, "tokens_seen": 93519872 }, { "epoch": 0.03, "learning_rate": 0.0004907321965897693, "loss": 3.3508, "theoretical_loss": 4.816509999129618, "tokens_seen": 93585408 }, { "epoch": 0.03, "learning_rate": 0.0004907221664994985, "loss": 3.5891, "theoretical_loss": 4.816038129891151, "tokens_seen": 93650944 }, { "epoch": 0.03, "learning_rate": 0.0004907121364092277, "loss": 3.4847, "theoretical_loss": 4.815566683131536, "tokens_seen": 93716480 }, { "epoch": 0.03, "learning_rate": 0.0004907021063189569, "loss": 3.7072, "theoretical_loss": 4.815095658177472, "tokens_seen": 93782016 }, { "epoch": 0.03, "learning_rate": 0.000490692076228686, "loss": 3.6021, "theoretical_loss": 4.814625054357199, "tokens_seen": 93847552 }, { "epoch": 0.03, "learning_rate": 0.0004906820461384152, "loss": 3.4324, "theoretical_loss": 4.814154871000497, "tokens_seen": 93913088 }, { "epoch": 0.03, "learning_rate": 0.0004906720160481444, "loss": 3.5133, "theoretical_loss": 4.813685107438679, "tokens_seen": 93978624 }, { "epoch": 0.03, "learning_rate": 0.0004906619859578736, "loss": 3.5597, "theoretical_loss": 4.813215763004585, "tokens_seen": 94044160 }, { "epoch": 0.03, "learning_rate": 0.0004906519558676028, "loss": 3.4239, "theoretical_loss": 4.812746837032582, "tokens_seen": 94109696 }, { "epoch": 0.03, "learning_rate": 0.0004906419257773321, "loss": 3.373, "theoretical_loss": 4.812278328858554, "tokens_seen": 94175232 }, { "epoch": 0.03, "learning_rate": 0.0004906318956870611, "loss": 3.5417, "theoretical_loss": 4.811810237819904, "tokens_seen": 94240768 }, { "epoch": 0.03, "learning_rate": 0.0004906218655967904, "loss": 3.3743, "theoretical_loss": 4.81134256325554, "tokens_seen": 94306304 }, { "epoch": 0.03, "learning_rate": 0.0004906118355065195, "loss": 3.5175, "theoretical_loss": 4.810875304505881, "tokens_seen": 94371840 }, { "epoch": 0.03, "learning_rate": 0.0004906018054162488, "loss": 3.3378, "theoretical_loss": 4.810408460912846, "tokens_seen": 94437376 }, { "epoch": 0.03, "learning_rate": 0.000490591775325978, "loss": 3.8392, "theoretical_loss": 4.809942031819853, "tokens_seen": 94502912 }, { "epoch": 0.03, "learning_rate": 0.0004905817452357072, "loss": 3.4628, "theoretical_loss": 4.809476016571809, "tokens_seen": 94568448 }, { "epoch": 0.03, "learning_rate": 0.0004905717151454363, "loss": 3.699, "theoretical_loss": 4.809010414515113, "tokens_seen": 94633984 }, { "epoch": 0.03, "learning_rate": 0.0004905616850551655, "loss": 3.6345, "theoretical_loss": 4.808545224997644, "tokens_seen": 94699520 }, { "epoch": 0.03, "learning_rate": 0.0004905516549648947, "loss": 3.3299, "theoretical_loss": 4.808080447368766, "tokens_seen": 94765056 }, { "epoch": 0.03, "learning_rate": 0.0004905416248746239, "loss": 3.5958, "theoretical_loss": 4.807616080979315, "tokens_seen": 94830592 }, { "epoch": 0.03, "learning_rate": 0.0004905315947843531, "loss": 3.752, "theoretical_loss": 4.807152125181597, "tokens_seen": 94896128 }, { "epoch": 0.03, "learning_rate": 0.0004905215646940822, "loss": 3.7867, "theoretical_loss": 4.806688579329387, "tokens_seen": 94961664 }, { "epoch": 0.03, "objective/train/docs_used": 65494, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3390982151031494, "objective/train/theoretical_loss": 4.8062254427779205, "objective/train/tokens_used": 115487200, "theoretical_loss": 4.8062254427779205, "tokens_seen": 95027200 }, { "epoch": 0.03, "learning_rate": 0.0004905115346038114, "loss": 3.5332, "theoretical_loss": 4.8062254427779205, "tokens_seen": 95027200 }, { "epoch": 0.03, "learning_rate": 0.0004905015045135406, "loss": 3.6865, "theoretical_loss": 4.80576271488389, "tokens_seen": 95092736 }, { "epoch": 0.03, "learning_rate": 0.0004904914744232698, "loss": 3.6984, "theoretical_loss": 4.805300395005444, "tokens_seen": 95158272 }, { "epoch": 0.03, "learning_rate": 0.000490481444332999, "loss": 3.7887, "theoretical_loss": 4.804838482502181, "tokens_seen": 95223808 }, { "epoch": 0.03, "learning_rate": 0.0004904714142427281, "loss": 3.5508, "theoretical_loss": 4.8043769767351385, "tokens_seen": 95289344 }, { "epoch": 0.03, "learning_rate": 0.0004904613841524574, "loss": 3.2765, "theoretical_loss": 4.8039158770668005, "tokens_seen": 95354880 }, { "epoch": 0.03, "learning_rate": 0.0004904513540621865, "loss": 3.4666, "theoretical_loss": 4.803455182861087, "tokens_seen": 95420416 }, { "epoch": 0.03, "learning_rate": 0.0004904413239719158, "loss": 3.4331, "theoretical_loss": 4.802994893483348, "tokens_seen": 95485952 }, { "epoch": 0.03, "learning_rate": 0.0004904312938816449, "loss": 3.4208, "theoretical_loss": 4.802535008300364, "tokens_seen": 95551488 }, { "epoch": 0.03, "learning_rate": 0.0004904212637913742, "loss": 3.3395, "theoretical_loss": 4.802075526680335, "tokens_seen": 95617024 }, { "epoch": 0.03, "learning_rate": 0.0004904112337011033, "loss": 3.3708, "theoretical_loss": 4.801616447992888, "tokens_seen": 95682560 }, { "epoch": 0.03, "learning_rate": 0.0004904012036108325, "loss": 3.6168, "theoretical_loss": 4.801157771609061, "tokens_seen": 95748096 }, { "epoch": 0.03, "learning_rate": 0.0004903911735205617, "loss": 3.3959, "theoretical_loss": 4.8006994969013, "tokens_seen": 95813632 }, { "epoch": 0.03, "learning_rate": 0.0004903811434302909, "loss": 3.6339, "theoretical_loss": 4.800241623243467, "tokens_seen": 95879168 }, { "epoch": 0.03, "learning_rate": 0.0004903711133400201, "loss": 3.3917, "theoretical_loss": 4.799784150010819, "tokens_seen": 95944704 }, { "epoch": 0.03, "learning_rate": 0.0004903610832497492, "loss": 3.6765, "theoretical_loss": 4.799327076580017, "tokens_seen": 96010240 }, { "epoch": 0.03, "learning_rate": 0.0004903510531594784, "loss": 3.6297, "theoretical_loss": 4.798870402329115, "tokens_seen": 96075776 }, { "epoch": 0.03, "learning_rate": 0.0004903410230692076, "loss": 3.4202, "theoretical_loss": 4.798414126637558, "tokens_seen": 96141312 }, { "epoch": 0.03, "learning_rate": 0.0004903309929789368, "loss": 3.417, "theoretical_loss": 4.797958248886179, "tokens_seen": 96206848 }, { "epoch": 0.03, "learning_rate": 0.000490320962888666, "loss": 3.5827, "theoretical_loss": 4.797502768457193, "tokens_seen": 96272384 }, { "epoch": 0.03, "learning_rate": 0.0004903109327983952, "loss": 3.5392, "theoretical_loss": 4.797047684734192, "tokens_seen": 96337920 }, { "epoch": 0.03, "learning_rate": 0.0004903009027081243, "loss": 3.744, "theoretical_loss": 4.796592997102147, "tokens_seen": 96403456 }, { "epoch": 0.03, "learning_rate": 0.0004902908726178535, "loss": 3.823, "theoretical_loss": 4.796138704947397, "tokens_seen": 96468992 }, { "epoch": 0.03, "learning_rate": 0.0004902808425275828, "loss": 3.475, "theoretical_loss": 4.795684807657649, "tokens_seen": 96534528 }, { "epoch": 0.03, "learning_rate": 0.0004902708124373119, "loss": 3.6812, "theoretical_loss": 4.795231304621968, "tokens_seen": 96600064 }, { "epoch": 0.03, "objective/train/docs_used": 66055, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.559401750564575, "objective/train/theoretical_loss": 4.794778195230787, "objective/train/tokens_used": 117125600, "theoretical_loss": 4.794778195230787, "tokens_seen": 96665600 }, { "epoch": 0.03, "learning_rate": 0.0004902607823470412, "loss": 3.6274, "theoretical_loss": 4.794778195230787, "tokens_seen": 96665600 }, { "epoch": 0.03, "learning_rate": 0.0004902507522567703, "loss": 3.5801, "theoretical_loss": 4.794325478875885, "tokens_seen": 96731136 }, { "epoch": 0.03, "learning_rate": 0.0004902407221664995, "loss": 3.6061, "theoretical_loss": 4.793873154950399, "tokens_seen": 96796672 }, { "epoch": 0.03, "learning_rate": 0.0004902306920762287, "loss": 3.6615, "theoretical_loss": 4.793421222848808, "tokens_seen": 96862208 }, { "epoch": 0.03, "learning_rate": 0.0004902206619859579, "loss": 3.3969, "theoretical_loss": 4.7929696819669365, "tokens_seen": 96927744 }, { "epoch": 0.03, "learning_rate": 0.0004902106318956871, "loss": 3.8594, "theoretical_loss": 4.792518531701948, "tokens_seen": 96993280 }, { "epoch": 0.03, "learning_rate": 0.0004902006018054163, "loss": 3.6387, "theoretical_loss": 4.792067771452341, "tokens_seen": 97058816 }, { "epoch": 0.03, "learning_rate": 0.0004901905717151454, "loss": 3.4593, "theoretical_loss": 4.791617400617948, "tokens_seen": 97124352 }, { "epoch": 0.03, "learning_rate": 0.0004901805416248746, "loss": 3.5499, "theoretical_loss": 4.791167418599925, "tokens_seen": 97189888 }, { "epoch": 0.03, "learning_rate": 0.0004901705115346038, "loss": 3.3165, "theoretical_loss": 4.790717824800755, "tokens_seen": 97255424 }, { "epoch": 0.03, "learning_rate": 0.000490160481444333, "loss": 3.5253, "theoretical_loss": 4.790268618624239, "tokens_seen": 97320960 }, { "epoch": 0.03, "learning_rate": 0.0004901504513540623, "loss": 3.2939, "theoretical_loss": 4.789819799475499, "tokens_seen": 97386496 }, { "epoch": 0.03, "learning_rate": 0.0004901404212637913, "loss": 3.7183, "theoretical_loss": 4.789371366760961, "tokens_seen": 97452032 }, { "epoch": 0.03, "learning_rate": 0.0004901303911735206, "loss": 3.3984, "theoretical_loss": 4.788923319888369, "tokens_seen": 97517568 }, { "epoch": 0.03, "learning_rate": 0.0004901203610832497, "loss": 3.4614, "theoretical_loss": 4.788475658266766, "tokens_seen": 97583104 }, { "epoch": 0.03, "learning_rate": 0.000490110330992979, "loss": 3.2819, "theoretical_loss": 4.788028381306497, "tokens_seen": 97648640 }, { "epoch": 0.03, "learning_rate": 0.0004901003009027082, "loss": 3.3862, "theoretical_loss": 4.787581488419207, "tokens_seen": 97714176 }, { "epoch": 0.03, "learning_rate": 0.0004900902708124374, "loss": 3.7962, "theoretical_loss": 4.787134979017832, "tokens_seen": 97779712 }, { "epoch": 0.03, "learning_rate": 0.0004900802407221665, "loss": 3.3423, "theoretical_loss": 4.786688852516599, "tokens_seen": 97845248 }, { "epoch": 0.03, "learning_rate": 0.0004900702106318957, "loss": 3.5008, "theoretical_loss": 4.786243108331024, "tokens_seen": 97910784 }, { "epoch": 0.03, "learning_rate": 0.0004900601805416249, "loss": 3.736, "theoretical_loss": 4.7857977458779, "tokens_seen": 97976320 }, { "epoch": 0.03, "learning_rate": 0.0004900501504513541, "loss": 3.6437, "theoretical_loss": 4.785352764575304, "tokens_seen": 98041856 }, { "epoch": 0.03, "learning_rate": 0.0004900401203610833, "loss": 3.4917, "theoretical_loss": 4.784908163842585, "tokens_seen": 98107392 }, { "epoch": 0.03, "learning_rate": 0.0004900300902708124, "loss": 3.6996, "theoretical_loss": 4.784463943100367, "tokens_seen": 98172928 }, { "epoch": 0.03, "learning_rate": 0.0004900200601805416, "loss": 3.5085, "theoretical_loss": 4.7840201017705395, "tokens_seen": 98238464 }, { "epoch": 0.03, "objective/train/docs_used": 66835, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 4.026576995849609, "objective/train/theoretical_loss": 4.783576639276257, "objective/train/tokens_used": 118764000, "theoretical_loss": 4.783576639276257, "tokens_seen": 98304000 }, { "epoch": 0.03, "learning_rate": 0.0004900100300902708, "loss": 3.6122, "theoretical_loss": 4.783576639276257, "tokens_seen": 98304000 }, { "epoch": 0.03, "learning_rate": 0.00049, "loss": 3.6488, "theoretical_loss": 4.783133555041934, "tokens_seen": 98369536 }, { "epoch": 0.03, "learning_rate": 0.0004899899699097292, "loss": 3.4521, "theoretical_loss": 4.782690848493245, "tokens_seen": 98435072 }, { "epoch": 0.03, "learning_rate": 0.0004899799398194583, "loss": 3.4946, "theoretical_loss": 4.7822485190571165, "tokens_seen": 98500608 }, { "epoch": 0.03, "learning_rate": 0.0004899699097291876, "loss": 3.1609, "theoretical_loss": 4.781806566161723, "tokens_seen": 98566144 }, { "epoch": 0.03, "learning_rate": 0.0004899598796389167, "loss": 3.2605, "theoretical_loss": 4.781364989236488, "tokens_seen": 98631680 }, { "epoch": 0.03, "learning_rate": 0.000489949849548646, "loss": 3.5803, "theoretical_loss": 4.78092378771208, "tokens_seen": 98697216 }, { "epoch": 0.03, "learning_rate": 0.0004899398194583751, "loss": 3.6634, "theoretical_loss": 4.780482961020402, "tokens_seen": 98762752 }, { "epoch": 0.03, "learning_rate": 0.0004899297893681044, "loss": 3.6734, "theoretical_loss": 4.780042508594596, "tokens_seen": 98828288 }, { "epoch": 0.03, "learning_rate": 0.0004899197592778335, "loss": 3.6452, "theoretical_loss": 4.779602429869035, "tokens_seen": 98893824 }, { "epoch": 0.03, "learning_rate": 0.0004899097291875627, "loss": 3.4789, "theoretical_loss": 4.779162724279324, "tokens_seen": 98959360 }, { "epoch": 0.03, "learning_rate": 0.0004898996990972919, "loss": 3.5272, "theoretical_loss": 4.7787233912622895, "tokens_seen": 99024896 }, { "epoch": 0.03, "learning_rate": 0.0004898896690070211, "loss": 3.9395, "theoretical_loss": 4.778284430255981, "tokens_seen": 99090432 }, { "epoch": 0.03, "learning_rate": 0.0004898796389167503, "loss": 3.803, "theoretical_loss": 4.77784584069967, "tokens_seen": 99155968 }, { "epoch": 0.03, "learning_rate": 0.0004898696088264794, "loss": 3.5448, "theoretical_loss": 4.777407622033838, "tokens_seen": 99221504 }, { "epoch": 0.03, "learning_rate": 0.0004898595787362086, "loss": 3.3819, "theoretical_loss": 4.776969773700181, "tokens_seen": 99287040 }, { "epoch": 0.03, "learning_rate": 0.0004898495486459378, "loss": 3.5351, "theoretical_loss": 4.776532295141601, "tokens_seen": 99352576 }, { "epoch": 0.03, "learning_rate": 0.000489839518555667, "loss": 3.6881, "theoretical_loss": 4.776095185802211, "tokens_seen": 99418112 }, { "epoch": 0.03, "learning_rate": 0.0004898294884653962, "loss": 3.8074, "theoretical_loss": 4.775658445127318, "tokens_seen": 99483648 }, { "epoch": 0.03, "learning_rate": 0.0004898194583751254, "loss": 3.4411, "theoretical_loss": 4.775222072563429, "tokens_seen": 99549184 }, { "epoch": 0.03, "learning_rate": 0.0004898094282848545, "loss": 3.4496, "theoretical_loss": 4.7747860675582485, "tokens_seen": 99614720 }, { "epoch": 0.03, "learning_rate": 0.0004897993981945837, "loss": 3.348, "theoretical_loss": 4.77435042956067, "tokens_seen": 99680256 }, { "epoch": 0.03, "learning_rate": 0.000489789368104313, "loss": 3.5864, "theoretical_loss": 4.773915158020776, "tokens_seen": 99745792 }, { "epoch": 0.03, "learning_rate": 0.0004897793380140421, "loss": 3.4012, "theoretical_loss": 4.773480252389831, "tokens_seen": 99811328 }, { "epoch": 0.03, "learning_rate": 0.0004897693079237714, "loss": 3.6203, "theoretical_loss": 4.773045712120284, "tokens_seen": 99876864 }, { "epoch": 0.03, "objective/train/docs_used": 67525, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.8278653621673584, "objective/train/theoretical_loss": 4.77261153666576, "objective/train/tokens_used": 120402400, "theoretical_loss": 4.77261153666576, "tokens_seen": 99942400 }, { "epoch": 0.03, "learning_rate": 0.0004897592778335005, "loss": 3.5655, "theoretical_loss": 4.77261153666576, "tokens_seen": 99942400 }, { "epoch": 0.03, "learning_rate": 0.0004897492477432297, "loss": 3.6504, "theoretical_loss": 4.772177725481062, "tokens_seen": 100007936 }, { "epoch": 0.03, "learning_rate": 0.0004897392176529589, "loss": 3.5848, "theoretical_loss": 4.77174427802216, "tokens_seen": 100073472 }, { "epoch": 0.03, "learning_rate": 0.0004897291875626881, "loss": 3.6098, "theoretical_loss": 4.771311193746191, "tokens_seen": 100139008 }, { "epoch": 0.03, "learning_rate": 0.0004897191574724173, "loss": 3.2068, "theoretical_loss": 4.770878472111465, "tokens_seen": 100204544 }, { "epoch": 0.03, "learning_rate": 0.0004897091273821465, "loss": 3.6295, "theoretical_loss": 4.770446112577445, "tokens_seen": 100270080 }, { "epoch": 0.03, "learning_rate": 0.0004896990972918756, "loss": 3.4111, "theoretical_loss": 4.770014114604756, "tokens_seen": 100335616 }, { "epoch": 0.03, "learning_rate": 0.0004896890672016048, "loss": 3.5571, "theoretical_loss": 4.769582477655177, "tokens_seen": 100401152 }, { "epoch": 0.03, "learning_rate": 0.000489679037111334, "loss": 3.2238, "theoretical_loss": 4.769151201191641, "tokens_seen": 100466688 }, { "epoch": 0.03, "learning_rate": 0.0004896690070210632, "loss": 3.501, "theoretical_loss": 4.768720284678228, "tokens_seen": 100532224 }, { "epoch": 0.03, "learning_rate": 0.0004896589769307924, "loss": 3.6858, "theoretical_loss": 4.768289727580161, "tokens_seen": 100597760 }, { "epoch": 0.03, "learning_rate": 0.0004896489468405215, "loss": 3.6342, "theoretical_loss": 4.767859529363809, "tokens_seen": 100663296 }, { "epoch": 0.03, "learning_rate": 0.0004896389167502507, "loss": 3.3918, "theoretical_loss": 4.767429689496682, "tokens_seen": 100728832 }, { "epoch": 0.03, "learning_rate": 0.0004896288866599799, "loss": 3.4626, "theoretical_loss": 4.767000207447417, "tokens_seen": 100794368 }, { "epoch": 0.03, "learning_rate": 0.0004896188565697091, "loss": 3.4291, "theoretical_loss": 4.766571082685794, "tokens_seen": 100859904 }, { "epoch": 0.03, "learning_rate": 0.0004896088264794384, "loss": 3.4262, "theoretical_loss": 4.766142314682716, "tokens_seen": 100925440 }, { "epoch": 0.03, "learning_rate": 0.0004895987963891674, "loss": 3.8865, "theoretical_loss": 4.765713902910214, "tokens_seen": 100990976 }, { "epoch": 0.03, "learning_rate": 0.0004895887662988967, "loss": 3.4615, "theoretical_loss": 4.765285846841444, "tokens_seen": 101056512 }, { "epoch": 0.03, "learning_rate": 0.0004895787362086259, "loss": 3.4272, "theoretical_loss": 4.76485814595068, "tokens_seen": 101122048 }, { "epoch": 0.03, "learning_rate": 0.0004895687061183551, "loss": 3.5119, "theoretical_loss": 4.764430799713314, "tokens_seen": 101187584 }, { "epoch": 0.03, "learning_rate": 0.0004895586760280843, "loss": 3.6724, "theoretical_loss": 4.764003807605853, "tokens_seen": 101253120 }, { "epoch": 0.03, "learning_rate": 0.0004895486459378135, "loss": 3.6045, "theoretical_loss": 4.763577169105912, "tokens_seen": 101318656 }, { "epoch": 0.03, "learning_rate": 0.0004895386158475426, "loss": 3.4985, "theoretical_loss": 4.763150883692218, "tokens_seen": 101384192 }, { "epoch": 0.03, "learning_rate": 0.0004895285857572718, "loss": 3.5738, "theoretical_loss": 4.762724950844598, "tokens_seen": 101449728 }, { "epoch": 0.03, "learning_rate": 0.000489518555667001, "loss": 3.4495, "theoretical_loss": 4.762299370043984, "tokens_seen": 101515264 }, { "epoch": 0.03, "objective/train/docs_used": 68617, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4468700885772705, "objective/train/theoretical_loss": 4.761874140772408, "objective/train/tokens_used": 122040800, "theoretical_loss": 4.761874140772408, "tokens_seen": 101580800 }, { "epoch": 0.03, "learning_rate": 0.0004895085255767302, "loss": 3.5015, "theoretical_loss": 4.761874140772408, "tokens_seen": 101580800 }, { "epoch": 0.03, "learning_rate": 0.0004894984954864594, "loss": 3.7421, "theoretical_loss": 4.761449262512993, "tokens_seen": 101646336 }, { "epoch": 0.03, "learning_rate": 0.0004894884653961885, "loss": 3.8129, "theoretical_loss": 4.761024734749958, "tokens_seen": 101711872 }, { "epoch": 0.03, "learning_rate": 0.0004894784353059178, "loss": 3.7874, "theoretical_loss": 4.76060055696861, "tokens_seen": 101777408 }, { "epoch": 0.03, "learning_rate": 0.0004894684052156469, "loss": 3.7841, "theoretical_loss": 4.760176728655345, "tokens_seen": 101842944 }, { "epoch": 0.03, "learning_rate": 0.0004894583751253762, "loss": 3.4924, "theoretical_loss": 4.75975324929764, "tokens_seen": 101908480 }, { "epoch": 0.03, "learning_rate": 0.0004894483450351053, "loss": 3.3683, "theoretical_loss": 4.759330118384053, "tokens_seen": 101974016 }, { "epoch": 0.03, "learning_rate": 0.0004894383149448346, "loss": 3.7427, "theoretical_loss": 4.758907335404221, "tokens_seen": 102039552 }, { "epoch": 0.03, "learning_rate": 0.0004894282848545637, "loss": 3.6284, "theoretical_loss": 4.758484899848854, "tokens_seen": 102105088 }, { "epoch": 0.03, "learning_rate": 0.0004894182547642929, "loss": 3.4244, "theoretical_loss": 4.7580628112097365, "tokens_seen": 102170624 }, { "epoch": 0.03, "learning_rate": 0.0004894082246740221, "loss": 3.4994, "theoretical_loss": 4.7576410689797175, "tokens_seen": 102236160 }, { "epoch": 0.03, "learning_rate": 0.0004893981945837513, "loss": 3.5959, "theoretical_loss": 4.757219672652717, "tokens_seen": 102301696 }, { "epoch": 0.03, "learning_rate": 0.0004893881644934805, "loss": 3.1601, "theoretical_loss": 4.756798621723712, "tokens_seen": 102367232 }, { "epoch": 0.03, "learning_rate": 0.0004893781344032096, "loss": 3.6548, "theoretical_loss": 4.756377915688748, "tokens_seen": 102432768 }, { "epoch": 0.03, "learning_rate": 0.0004893681043129388, "loss": 3.4761, "theoretical_loss": 4.755957554044917, "tokens_seen": 102498304 }, { "epoch": 0.03, "learning_rate": 0.000489358074222668, "loss": 3.5187, "theoretical_loss": 4.755537536290373, "tokens_seen": 102563840 }, { "epoch": 0.03, "learning_rate": 0.0004893480441323972, "loss": 3.4874, "theoretical_loss": 4.755117861924321, "tokens_seen": 102629376 }, { "epoch": 0.03, "learning_rate": 0.0004893380140421264, "loss": 3.387, "theoretical_loss": 4.754698530447009, "tokens_seen": 102694912 }, { "epoch": 0.03, "learning_rate": 0.0004893279839518556, "loss": 3.3774, "theoretical_loss": 4.754279541359738, "tokens_seen": 102760448 }, { "epoch": 0.03, "learning_rate": 0.0004893179538615847, "loss": 3.4792, "theoretical_loss": 4.753860894164845, "tokens_seen": 102825984 }, { "epoch": 0.03, "learning_rate": 0.0004893079237713139, "loss": 3.5093, "theoretical_loss": 4.75344258836571, "tokens_seen": 102891520 }, { "epoch": 0.03, "learning_rate": 0.0004892978936810432, "loss": 3.5069, "theoretical_loss": 4.753024623466752, "tokens_seen": 102957056 }, { "epoch": 0.03, "learning_rate": 0.0004892878635907723, "loss": 3.7293, "theoretical_loss": 4.752606998973421, "tokens_seen": 103022592 }, { "epoch": 0.03, "learning_rate": 0.0004892778335005016, "loss": 3.5777, "theoretical_loss": 4.752189714392202, "tokens_seen": 103088128 }, { "epoch": 0.03, "learning_rate": 0.0004892678034102307, "loss": 3.5236, "theoretical_loss": 4.7517727692306035, "tokens_seen": 103153664 }, { "epoch": 0.03, "objective/train/docs_used": 69265, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0723447799682617, "objective/train/theoretical_loss": 4.751356162997164, "objective/train/tokens_used": 123679200, "theoretical_loss": 4.751356162997164, "tokens_seen": 103219200 }, { "epoch": 0.03, "learning_rate": 0.0004892577733199599, "loss": 3.525, "theoretical_loss": 4.751356162997164, "tokens_seen": 103219200 }, { "epoch": 0.03, "learning_rate": 0.0004892477432296891, "loss": 3.602, "theoretical_loss": 4.750939895201443, "tokens_seen": 103284736 }, { "epoch": 0.03, "learning_rate": 0.0004892377131394183, "loss": 3.4738, "theoretical_loss": 4.750523965354024, "tokens_seen": 103350272 }, { "epoch": 0.03, "learning_rate": 0.0004892276830491475, "loss": 3.3909, "theoretical_loss": 4.750108372966501, "tokens_seen": 103415808 }, { "epoch": 0.03, "learning_rate": 0.0004892176529588767, "loss": 3.2759, "theoretical_loss": 4.749693117551491, "tokens_seen": 103481344 }, { "epoch": 0.03, "learning_rate": 0.0004892076228686058, "loss": 3.7655, "theoretical_loss": 4.749278198622617, "tokens_seen": 103546880 }, { "epoch": 0.03, "learning_rate": 0.000489197592778335, "loss": 3.4825, "theoretical_loss": 4.748863615694514, "tokens_seen": 103612416 }, { "epoch": 0.03, "learning_rate": 0.0004891875626880642, "loss": 3.7804, "theoretical_loss": 4.748449368282822, "tokens_seen": 103677952 }, { "epoch": 0.03, "learning_rate": 0.0004891775325977934, "loss": 3.4161, "theoretical_loss": 4.748035455904185, "tokens_seen": 103743488 }, { "epoch": 0.03, "learning_rate": 0.0004891675025075226, "loss": 3.5973, "theoretical_loss": 4.747621878076252, "tokens_seen": 103809024 }, { "epoch": 0.03, "learning_rate": 0.0004891574724172517, "loss": 3.5654, "theoretical_loss": 4.747208634317664, "tokens_seen": 103874560 }, { "epoch": 0.03, "learning_rate": 0.0004891474423269809, "loss": 3.6798, "theoretical_loss": 4.746795724148061, "tokens_seen": 103940096 }, { "epoch": 0.03, "learning_rate": 0.0004891374122367101, "loss": 3.7902, "theoretical_loss": 4.746383147088078, "tokens_seen": 104005632 }, { "epoch": 0.03, "learning_rate": 0.0004891273821464393, "loss": 3.2944, "theoretical_loss": 4.745970902659338, "tokens_seen": 104071168 }, { "epoch": 0.03, "learning_rate": 0.0004891173520561686, "loss": 3.6076, "theoretical_loss": 4.745558990384451, "tokens_seen": 104136704 }, { "epoch": 0.03, "learning_rate": 0.0004891073219658976, "loss": 3.3813, "theoretical_loss": 4.7451474097870125, "tokens_seen": 104202240 }, { "epoch": 0.03, "learning_rate": 0.0004890972918756269, "loss": 3.245, "theoretical_loss": 4.744736160391602, "tokens_seen": 104267776 }, { "epoch": 0.03, "learning_rate": 0.0004890872617853561, "loss": 3.6455, "theoretical_loss": 4.744325241723777, "tokens_seen": 104333312 }, { "epoch": 0.03, "learning_rate": 0.0004890772316950853, "loss": 3.6057, "theoretical_loss": 4.743914653310073, "tokens_seen": 104398848 }, { "epoch": 0.03, "learning_rate": 0.0004890672016048145, "loss": 3.2381, "theoretical_loss": 4.743504394678, "tokens_seen": 104464384 }, { "epoch": 0.03, "learning_rate": 0.0004890571715145437, "loss": 3.5268, "theoretical_loss": 4.743094465356039, "tokens_seen": 104529920 }, { "epoch": 0.03, "learning_rate": 0.0004890471414242728, "loss": 3.4362, "theoretical_loss": 4.742684864873641, "tokens_seen": 104595456 }, { "epoch": 0.03, "learning_rate": 0.000489037111334002, "loss": 3.2982, "theoretical_loss": 4.742275592761223, "tokens_seen": 104660992 }, { "epoch": 0.03, "learning_rate": 0.0004890270812437312, "loss": 3.5981, "theoretical_loss": 4.741866648550168, "tokens_seen": 104726528 }, { "epoch": 0.03, "learning_rate": 0.0004890170511534604, "loss": 3.5663, "theoretical_loss": 4.741458031772817, "tokens_seen": 104792064 }, { "epoch": 0.03, "objective/train/docs_used": 70556, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0196893215179443, "objective/train/theoretical_loss": 4.741049741962473, "objective/train/tokens_used": 125317600, "theoretical_loss": 4.741049741962473, "tokens_seen": 104857600 }, { "epoch": 0.03, "learning_rate": 0.0004890070210631896, "loss": 3.3116, "theoretical_loss": 4.741049741962473, "tokens_seen": 104857600 }, { "epoch": 0.03, "learning_rate": 0.0004889969909729187, "loss": 3.3268, "theoretical_loss": 4.740641778653395, "tokens_seen": 104923136 }, { "epoch": 0.03, "learning_rate": 0.0004889869608826479, "loss": 3.535, "theoretical_loss": 4.740234141380794, "tokens_seen": 104988672 }, { "epoch": 0.03, "learning_rate": 0.0004889769307923771, "loss": 3.5609, "theoretical_loss": 4.739826829680833, "tokens_seen": 105054208 }, { "epoch": 0.03, "learning_rate": 0.0004889669007021063, "loss": 3.2398, "theoretical_loss": 4.739419843090626, "tokens_seen": 105119744 }, { "epoch": 0.03, "learning_rate": 0.0004889568706118355, "loss": 3.6536, "theoretical_loss": 4.739013181148229, "tokens_seen": 105185280 }, { "epoch": 0.03, "learning_rate": 0.0004889468405215647, "loss": 3.2403, "theoretical_loss": 4.738606843392644, "tokens_seen": 105250816 }, { "epoch": 0.03, "learning_rate": 0.0004889368104312939, "loss": 3.3196, "theoretical_loss": 4.738200829363815, "tokens_seen": 105316352 }, { "epoch": 0.03, "learning_rate": 0.000488926780341023, "loss": 3.5581, "theoretical_loss": 4.737795138602624, "tokens_seen": 105381888 }, { "epoch": 0.03, "learning_rate": 0.0004889167502507523, "loss": 3.478, "theoretical_loss": 4.737389770650887, "tokens_seen": 105447424 }, { "epoch": 0.03, "learning_rate": 0.0004889067201604815, "loss": 3.4703, "theoretical_loss": 4.736984725051357, "tokens_seen": 105512960 }, { "epoch": 0.03, "learning_rate": 0.0004888966900702107, "loss": 3.5458, "theoretical_loss": 4.736580001347717, "tokens_seen": 105578496 }, { "epoch": 0.03, "learning_rate": 0.0004888866599799398, "loss": 3.5523, "theoretical_loss": 4.736175599084576, "tokens_seen": 105644032 }, { "epoch": 0.03, "learning_rate": 0.000488876629889669, "loss": 3.4106, "theoretical_loss": 4.735771517807473, "tokens_seen": 105709568 }, { "epoch": 0.03, "learning_rate": 0.0004888665997993982, "loss": 3.6541, "theoretical_loss": 4.735367757062869, "tokens_seen": 105775104 }, { "epoch": 0.03, "learning_rate": 0.0004888565697091274, "loss": 3.4943, "theoretical_loss": 4.734964316398148, "tokens_seen": 105840640 }, { "epoch": 0.03, "learning_rate": 0.0004888465396188566, "loss": 3.4438, "theoretical_loss": 4.734561195361609, "tokens_seen": 105906176 }, { "epoch": 0.03, "learning_rate": 0.0004888365095285858, "loss": 3.5359, "theoretical_loss": 4.734158393502471, "tokens_seen": 105971712 }, { "epoch": 0.03, "learning_rate": 0.0004888264794383149, "loss": 3.6042, "theoretical_loss": 4.733755910370867, "tokens_seen": 106037248 }, { "epoch": 0.03, "learning_rate": 0.0004888164493480441, "loss": 3.4694, "theoretical_loss": 4.73335374551784, "tokens_seen": 106102784 }, { "epoch": 0.03, "learning_rate": 0.0004888064192577733, "loss": 3.3701, "theoretical_loss": 4.732951898495341, "tokens_seen": 106168320 }, { "epoch": 0.03, "learning_rate": 0.0004887963891675025, "loss": 3.4071, "theoretical_loss": 4.7325503688562325, "tokens_seen": 106233856 }, { "epoch": 0.03, "learning_rate": 0.0004887863590772317, "loss": 3.4026, "theoretical_loss": 4.732149156154276, "tokens_seen": 106299392 }, { "epoch": 0.03, "learning_rate": 0.000488776328986961, "loss": 3.5821, "theoretical_loss": 4.731748259944139, "tokens_seen": 106364928 }, { "epoch": 0.03, "learning_rate": 0.00048876629889669, "loss": 3.1001, "theoretical_loss": 4.731347679781386, "tokens_seen": 106430464 }, { "epoch": 0.03, "objective/train/docs_used": 70950, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.365251302719116, "objective/train/theoretical_loss": 4.730947415222481, "objective/train/tokens_used": 126956000, "theoretical_loss": 4.730947415222481, "tokens_seen": 106496000 }, { "epoch": 0.03, "learning_rate": 0.0004887562688064193, "loss": 3.3721, "theoretical_loss": 4.730947415222481, "tokens_seen": 106496000 }, { "epoch": 0.03, "learning_rate": 0.0004887462387161484, "loss": 3.3913, "theoretical_loss": 4.730547465824781, "tokens_seen": 106561536 }, { "epoch": 0.03, "learning_rate": 0.0004887362086258777, "loss": 3.477, "theoretical_loss": 4.730147831146537, "tokens_seen": 106627072 }, { "epoch": 0.03, "learning_rate": 0.0004887261785356069, "loss": 3.4452, "theoretical_loss": 4.72974851074689, "tokens_seen": 106692608 }, { "epoch": 0.03, "learning_rate": 0.000488716148445336, "loss": 3.3808, "theoretical_loss": 4.729349504185867, "tokens_seen": 106758144 }, { "epoch": 0.03, "learning_rate": 0.0004887061183550652, "loss": 3.6197, "theoretical_loss": 4.728950811024383, "tokens_seen": 106823680 }, { "epoch": 0.03, "learning_rate": 0.0004886960882647944, "loss": 3.2728, "theoretical_loss": 4.7285524308242355, "tokens_seen": 106889216 }, { "epoch": 0.03, "learning_rate": 0.0004886860581745236, "loss": 3.4828, "theoretical_loss": 4.728154363148102, "tokens_seen": 106954752 }, { "epoch": 0.03, "learning_rate": 0.0004886760280842528, "loss": 3.645, "theoretical_loss": 4.72775660755954, "tokens_seen": 107020288 }, { "epoch": 0.03, "learning_rate": 0.0004886659979939819, "loss": 3.5389, "theoretical_loss": 4.72735916362298, "tokens_seen": 107085824 }, { "epoch": 0.03, "learning_rate": 0.0004886559679037111, "loss": 3.2972, "theoretical_loss": 4.7269620309037315, "tokens_seen": 107151360 }, { "epoch": 0.03, "learning_rate": 0.0004886459378134403, "loss": 3.3874, "theoretical_loss": 4.726565208967973, "tokens_seen": 107216896 }, { "epoch": 0.03, "learning_rate": 0.0004886359077231695, "loss": 3.375, "theoretical_loss": 4.726168697382751, "tokens_seen": 107282432 }, { "epoch": 0.03, "learning_rate": 0.0004886258776328988, "loss": 3.6601, "theoretical_loss": 4.725772495715983, "tokens_seen": 107347968 }, { "epoch": 0.03, "learning_rate": 0.0004886158475426278, "loss": 3.369, "theoretical_loss": 4.725376603536446, "tokens_seen": 107413504 }, { "epoch": 0.03, "learning_rate": 0.0004886058174523571, "loss": 3.2596, "theoretical_loss": 4.724981020413787, "tokens_seen": 107479040 }, { "epoch": 0.03, "learning_rate": 0.0004885957873620863, "loss": 3.3279, "theoretical_loss": 4.724585745918505, "tokens_seen": 107544576 }, { "epoch": 0.03, "learning_rate": 0.0004885857572718155, "loss": 3.4956, "theoretical_loss": 4.7241907796219635, "tokens_seen": 107610112 }, { "epoch": 0.03, "learning_rate": 0.0004885757271815447, "loss": 3.4726, "theoretical_loss": 4.723796121096381, "tokens_seen": 107675648 }, { "epoch": 0.03, "learning_rate": 0.0004885656970912739, "loss": 3.5414, "theoretical_loss": 4.723401769914824, "tokens_seen": 107741184 }, { "epoch": 0.03, "learning_rate": 0.000488555667001003, "loss": 3.2837, "theoretical_loss": 4.723007725651219, "tokens_seen": 107806720 }, { "epoch": 0.03, "learning_rate": 0.0004885456369107322, "loss": 3.4393, "theoretical_loss": 4.722613987880335, "tokens_seen": 107872256 }, { "epoch": 0.03, "learning_rate": 0.0004885356068204614, "loss": 3.4046, "theoretical_loss": 4.722220556177792, "tokens_seen": 107937792 }, { "epoch": 0.03, "learning_rate": 0.0004885255767301906, "loss": 3.3551, "theoretical_loss": 4.721827430120053, "tokens_seen": 108003328 }, { "epoch": 0.03, "learning_rate": 0.0004885155466399198, "loss": 3.4638, "theoretical_loss": 4.721434609284424, "tokens_seen": 108068864 }, { "epoch": 0.03, "objective/train/docs_used": 72030, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6380910873413086, "objective/train/theoretical_loss": 4.721042093249051, "objective/train/tokens_used": 128594400, "theoretical_loss": 4.721042093249051, "tokens_seen": 108134400 }, { "epoch": 0.03, "learning_rate": 0.000488505516549649, "loss": 3.2402, "theoretical_loss": 4.721042093249051, "tokens_seen": 108134400 }, { "epoch": 0.03, "learning_rate": 0.0004884954864593781, "loss": 3.3689, "theoretical_loss": 4.720649881592919, "tokens_seen": 108199936 }, { "epoch": 0.03, "learning_rate": 0.0004884854563691073, "loss": 3.5546, "theoretical_loss": 4.7202579738958494, "tokens_seen": 108265472 }, { "epoch": 0.03, "learning_rate": 0.0004884754262788365, "loss": 3.2907, "theoretical_loss": 4.7198663697384955, "tokens_seen": 108331008 }, { "epoch": 0.03, "learning_rate": 0.0004884653961885657, "loss": 3.763, "theoretical_loss": 4.719475068702346, "tokens_seen": 108396544 }, { "epoch": 0.03, "learning_rate": 0.0004884553660982949, "loss": 3.3741, "theoretical_loss": 4.719084070369714, "tokens_seen": 108462080 }, { "epoch": 0.03, "learning_rate": 0.0004884453360080241, "loss": 3.5399, "theoretical_loss": 4.718693374323747, "tokens_seen": 108527616 }, { "epoch": 0.03, "learning_rate": 0.0004884353059177532, "loss": 3.5136, "theoretical_loss": 4.718302980148412, "tokens_seen": 108593152 }, { "epoch": 0.03, "learning_rate": 0.0004884252758274825, "loss": 3.3874, "theoretical_loss": 4.717912887428501, "tokens_seen": 108658688 }, { "epoch": 0.03, "learning_rate": 0.0004884152457372117, "loss": 3.2129, "theoretical_loss": 4.717523095749626, "tokens_seen": 108724224 }, { "epoch": 0.03, "learning_rate": 0.0004884052156469409, "loss": 3.1945, "theoretical_loss": 4.717133604698222, "tokens_seen": 108789760 }, { "epoch": 0.03, "learning_rate": 0.00048839518555667, "loss": 3.4207, "theoretical_loss": 4.7167444138615355, "tokens_seen": 108855296 }, { "epoch": 0.03, "learning_rate": 0.0004883851554663992, "loss": 3.6135, "theoretical_loss": 4.716355522827633, "tokens_seen": 108920832 }, { "epoch": 0.03, "learning_rate": 0.0004883751253761284, "loss": 3.7415, "theoretical_loss": 4.715966931185388, "tokens_seen": 108986368 }, { "epoch": 0.03, "learning_rate": 0.0004883650952858576, "loss": 3.6946, "theoretical_loss": 4.715578638524491, "tokens_seen": 109051904 }, { "epoch": 0.03, "learning_rate": 0.0004883550651955868, "loss": 3.4421, "theoretical_loss": 4.715190644435435, "tokens_seen": 109117440 }, { "epoch": 0.03, "learning_rate": 0.000488345035105316, "loss": 3.6584, "theoretical_loss": 4.714802948509522, "tokens_seen": 109182976 }, { "epoch": 0.03, "learning_rate": 0.0004883350050150451, "loss": 3.3107, "theoretical_loss": 4.71441555033886, "tokens_seen": 109248512 }, { "epoch": 0.03, "learning_rate": 0.0004883249749247743, "loss": 3.5758, "theoretical_loss": 4.714028449516356, "tokens_seen": 109314048 }, { "epoch": 0.03, "learning_rate": 0.0004883149448345035, "loss": 3.3739, "theoretical_loss": 4.713641645635718, "tokens_seen": 109379584 }, { "epoch": 0.03, "learning_rate": 0.0004883049147442327, "loss": 3.1408, "theoretical_loss": 4.713255138291454, "tokens_seen": 109445120 }, { "epoch": 0.03, "learning_rate": 0.0004882948846539619, "loss": 3.6864, "theoretical_loss": 4.712868927078868, "tokens_seen": 109510656 }, { "epoch": 0.03, "learning_rate": 0.0004882848545636911, "loss": 3.2615, "theoretical_loss": 4.712483011594056, "tokens_seen": 109576192 }, { "epoch": 0.03, "learning_rate": 0.0004882748244734203, "loss": 3.5947, "theoretical_loss": 4.7120973914339075, "tokens_seen": 109641728 }, { "epoch": 0.03, "learning_rate": 0.00048826479438314946, "loss": 3.3371, "theoretical_loss": 4.7117120661961005, "tokens_seen": 109707264 }, { "epoch": 0.03, "objective/train/docs_used": 72698, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.991194009780884, "objective/train/theoretical_loss": 4.711327035479103, "objective/train/tokens_used": 130232800, "theoretical_loss": 4.711327035479103, "tokens_seen": 109772800 }, { "epoch": 0.03, "learning_rate": 0.00048825476429287864, "loss": 3.2689, "theoretical_loss": 4.711327035479103, "tokens_seen": 109772800 }, { "epoch": 0.03, "learning_rate": 0.0004882447342026078, "loss": 3.4525, "theoretical_loss": 4.710942298882169, "tokens_seen": 109838336 }, { "epoch": 0.03, "learning_rate": 0.000488234704112337, "loss": 3.3659, "theoretical_loss": 4.710557856005335, "tokens_seen": 109903872 }, { "epoch": 0.03, "learning_rate": 0.00048822467402206624, "loss": 3.4411, "theoretical_loss": 4.710173706449419, "tokens_seen": 109969408 }, { "epoch": 0.03, "learning_rate": 0.00048821464393179536, "loss": 3.382, "theoretical_loss": 4.709789849816021, "tokens_seen": 110034944 }, { "epoch": 0.03, "learning_rate": 0.0004882046138415246, "loss": 3.3798, "theoretical_loss": 4.7094062857075185, "tokens_seen": 110100480 }, { "epoch": 0.03, "learning_rate": 0.0004881945837512537, "loss": 3.5293, "theoretical_loss": 4.709023013727063, "tokens_seen": 110166016 }, { "epoch": 0.03, "learning_rate": 0.00048818455366098296, "loss": 3.6122, "theoretical_loss": 4.708640033478584, "tokens_seen": 110231552 }, { "epoch": 0.03, "learning_rate": 0.00048817452357071214, "loss": 3.3274, "theoretical_loss": 4.708257344566778, "tokens_seen": 110297088 }, { "epoch": 0.03, "learning_rate": 0.0004881644934804413, "loss": 3.554, "theoretical_loss": 4.7078749465971175, "tokens_seen": 110362624 }, { "epoch": 0.03, "learning_rate": 0.0004881544633901705, "loss": 3.3524, "theoretical_loss": 4.707492839175837, "tokens_seen": 110428160 }, { "epoch": 0.03, "learning_rate": 0.00048814443329989974, "loss": 3.5396, "theoretical_loss": 4.707111021909941, "tokens_seen": 110493696 }, { "epoch": 0.03, "learning_rate": 0.00048813440320962887, "loss": 3.5515, "theoretical_loss": 4.706729494407197, "tokens_seen": 110559232 }, { "epoch": 0.03, "learning_rate": 0.0004881243731193581, "loss": 3.2664, "theoretical_loss": 4.706348256276138, "tokens_seen": 110624768 }, { "epoch": 0.03, "learning_rate": 0.00048811434302908723, "loss": 3.4359, "theoretical_loss": 4.705967307126051, "tokens_seen": 110690304 }, { "epoch": 0.03, "learning_rate": 0.00048810431293881646, "loss": 3.6256, "theoretical_loss": 4.705586646566987, "tokens_seen": 110755840 }, { "epoch": 0.03, "learning_rate": 0.00048809428284854564, "loss": 3.5248, "theoretical_loss": 4.705206274209751, "tokens_seen": 110821376 }, { "epoch": 0.03, "learning_rate": 0.0004880842527582748, "loss": 3.5576, "theoretical_loss": 4.704826189665905, "tokens_seen": 110886912 }, { "epoch": 0.03, "learning_rate": 0.000488074222668004, "loss": 3.3691, "theoretical_loss": 4.704446392547759, "tokens_seen": 110952448 }, { "epoch": 0.03, "learning_rate": 0.0004880641925777332, "loss": 3.6296, "theoretical_loss": 4.7040668824683785, "tokens_seen": 111017984 }, { "epoch": 0.03, "learning_rate": 0.00048805416248746237, "loss": 3.5711, "theoretical_loss": 4.7036876590415755, "tokens_seen": 111083520 }, { "epoch": 0.03, "learning_rate": 0.0004880441323971916, "loss": 3.1637, "theoretical_loss": 4.7033087218819105, "tokens_seen": 111149056 }, { "epoch": 0.03, "learning_rate": 0.00048803410230692073, "loss": 3.4891, "theoretical_loss": 4.7029300706046895, "tokens_seen": 111214592 }, { "epoch": 0.03, "learning_rate": 0.00048802407221664997, "loss": 3.4164, "theoretical_loss": 4.702551704825957, "tokens_seen": 111280128 }, { "epoch": 0.03, "learning_rate": 0.00048801404212637915, "loss": 3.2079, "theoretical_loss": 4.702173624162507, "tokens_seen": 111345664 }, { "epoch": 0.03, "objective/train/docs_used": 74080, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.588409900665283, "objective/train/theoretical_loss": 4.701795828231866, "objective/train/tokens_used": 131871200, "theoretical_loss": 4.701795828231866, "tokens_seen": 111411200 }, { "epoch": 0.03, "learning_rate": 0.00048800401203610833, "loss": 3.4279, "theoretical_loss": 4.701795828231866, "tokens_seen": 111411200 }, { "epoch": 0.03, "learning_rate": 0.0004879939819458375, "loss": 3.4107, "theoretical_loss": 4.701418316652299, "tokens_seen": 111476736 }, { "epoch": 0.03, "learning_rate": 0.0004879839518555667, "loss": 3.4536, "theoretical_loss": 4.701041089042813, "tokens_seen": 111542272 }, { "epoch": 0.03, "learning_rate": 0.0004879739217652959, "loss": 3.7348, "theoretical_loss": 4.700664145023142, "tokens_seen": 111607808 }, { "epoch": 0.03, "learning_rate": 0.0004879638916750251, "loss": 3.4685, "theoretical_loss": 4.700287484213753, "tokens_seen": 111673344 }, { "epoch": 0.03, "learning_rate": 0.0004879538615847543, "loss": 3.4494, "theoretical_loss": 4.699911106235849, "tokens_seen": 111738880 }, { "epoch": 0.03, "learning_rate": 0.00048794383149448347, "loss": 3.5418, "theoretical_loss": 4.6995350107113545, "tokens_seen": 111804416 }, { "epoch": 0.03, "learning_rate": 0.00048793380140421265, "loss": 3.5618, "theoretical_loss": 4.699159197262922, "tokens_seen": 111869952 }, { "epoch": 0.03, "learning_rate": 0.00048792377131394183, "loss": 3.3049, "theoretical_loss": 4.698783665513934, "tokens_seen": 111935488 }, { "epoch": 0.03, "learning_rate": 0.00048791374122367107, "loss": 3.4054, "theoretical_loss": 4.698408415088491, "tokens_seen": 112001024 }, { "epoch": 0.03, "learning_rate": 0.0004879037111334002, "loss": 3.5376, "theoretical_loss": 4.698033445611415, "tokens_seen": 112066560 }, { "epoch": 0.03, "learning_rate": 0.00048789368104312943, "loss": 3.5404, "theoretical_loss": 4.6976587567082495, "tokens_seen": 112132096 }, { "epoch": 0.03, "learning_rate": 0.00048788365095285856, "loss": 3.6666, "theoretical_loss": 4.697284348005253, "tokens_seen": 112197632 }, { "epoch": 0.03, "learning_rate": 0.0004878736208625878, "loss": 3.6087, "theoretical_loss": 4.696910219129402, "tokens_seen": 112263168 }, { "epoch": 0.03, "learning_rate": 0.00048786359077231697, "loss": 3.5387, "theoretical_loss": 4.696536369708386, "tokens_seen": 112328704 }, { "epoch": 0.03, "learning_rate": 0.00048785356068204615, "loss": 3.3868, "theoretical_loss": 4.696162799370606, "tokens_seen": 112394240 }, { "epoch": 0.03, "learning_rate": 0.00048784353059177533, "loss": 3.6141, "theoretical_loss": 4.695789507745176, "tokens_seen": 112459776 }, { "epoch": 0.03, "learning_rate": 0.00048783350050150457, "loss": 3.5212, "theoretical_loss": 4.695416494461917, "tokens_seen": 112525312 }, { "epoch": 0.03, "learning_rate": 0.0004878234704112337, "loss": 3.4791, "theoretical_loss": 4.695043759151353, "tokens_seen": 112590848 }, { "epoch": 0.03, "learning_rate": 0.00048781344032096293, "loss": 3.4949, "theoretical_loss": 4.694671301444722, "tokens_seen": 112656384 }, { "epoch": 0.03, "learning_rate": 0.00048780341023069206, "loss": 3.6483, "theoretical_loss": 4.694299120973957, "tokens_seen": 112721920 }, { "epoch": 0.03, "learning_rate": 0.0004877933801404213, "loss": 3.5765, "theoretical_loss": 4.693927217371698, "tokens_seen": 112787456 }, { "epoch": 0.03, "learning_rate": 0.0004877833500501505, "loss": 3.5005, "theoretical_loss": 4.693555590271282, "tokens_seen": 112852992 }, { "epoch": 0.03, "learning_rate": 0.00048777331995987966, "loss": 3.3525, "theoretical_loss": 4.693184239306744, "tokens_seen": 112918528 }, { "epoch": 0.03, "learning_rate": 0.00048776328986960884, "loss": 3.5055, "theoretical_loss": 4.692813164112819, "tokens_seen": 112984064 }, { "epoch": 0.03, "objective/train/docs_used": 74817, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.344128131866455, "objective/train/theoretical_loss": 4.692442364324931, "objective/train/tokens_used": 133509600, "theoretical_loss": 4.692442364324931, "tokens_seen": 113049600 }, { "epoch": 0.03, "learning_rate": 0.000487753259779338, "loss": 3.3455, "theoretical_loss": 4.692442364324931, "tokens_seen": 113049600 }, { "epoch": 0.03, "learning_rate": 0.0004877432296890672, "loss": 3.4303, "theoretical_loss": 4.692071839579201, "tokens_seen": 113115136 }, { "epoch": 0.03, "learning_rate": 0.00048773319959879644, "loss": 3.4279, "theoretical_loss": 4.6917015895124425, "tokens_seen": 113180672 }, { "epoch": 0.03, "learning_rate": 0.00048772316950852556, "loss": 3.7147, "theoretical_loss": 4.691331613762153, "tokens_seen": 113246208 }, { "epoch": 0.03, "learning_rate": 0.0004877131394182548, "loss": 3.4796, "theoretical_loss": 4.690961911966523, "tokens_seen": 113311744 }, { "epoch": 0.03, "learning_rate": 0.0004877031093279839, "loss": 3.4674, "theoretical_loss": 4.690592483764427, "tokens_seen": 113377280 }, { "epoch": 0.03, "learning_rate": 0.00048769307923771316, "loss": 3.4433, "theoretical_loss": 4.690223328795424, "tokens_seen": 113442816 }, { "epoch": 0.03, "learning_rate": 0.00048768304914744234, "loss": 3.2091, "theoretical_loss": 4.689854446699757, "tokens_seen": 113508352 }, { "epoch": 0.03, "learning_rate": 0.0004876730190571715, "loss": 3.345, "theoretical_loss": 4.689485837118347, "tokens_seen": 113573888 }, { "epoch": 0.03, "learning_rate": 0.0004876629889669007, "loss": 3.3691, "theoretical_loss": 4.689117499692798, "tokens_seen": 113639424 }, { "epoch": 0.03, "learning_rate": 0.00048765295887662994, "loss": 3.5643, "theoretical_loss": 4.688749434065389, "tokens_seen": 113704960 }, { "epoch": 0.03, "learning_rate": 0.00048764292878635907, "loss": 3.241, "theoretical_loss": 4.688381639879076, "tokens_seen": 113770496 }, { "epoch": 0.03, "learning_rate": 0.0004876328986960883, "loss": 3.4272, "theoretical_loss": 4.68801411677749, "tokens_seen": 113836032 }, { "epoch": 0.03, "learning_rate": 0.00048762286860581743, "loss": 3.3181, "theoretical_loss": 4.687646864404934, "tokens_seen": 113901568 }, { "epoch": 0.03, "learning_rate": 0.00048761283851554666, "loss": 3.6051, "theoretical_loss": 4.687279882406381, "tokens_seen": 113967104 }, { "epoch": 0.03, "learning_rate": 0.00048760280842527584, "loss": 3.1655, "theoretical_loss": 4.686913170427477, "tokens_seen": 114032640 }, { "epoch": 0.03, "learning_rate": 0.000487592778335005, "loss": 3.3195, "theoretical_loss": 4.68654672811453, "tokens_seen": 114098176 }, { "epoch": 0.03, "learning_rate": 0.0004875827482447342, "loss": 3.4204, "theoretical_loss": 4.68618055511452, "tokens_seen": 114163712 }, { "epoch": 0.03, "learning_rate": 0.0004875727181544634, "loss": 3.3986, "theoretical_loss": 4.685814651075088, "tokens_seen": 114229248 }, { "epoch": 0.03, "learning_rate": 0.00048756268806419257, "loss": 3.3853, "theoretical_loss": 4.685449015644537, "tokens_seen": 114294784 }, { "epoch": 0.03, "learning_rate": 0.0004875526579739218, "loss": 3.1979, "theoretical_loss": 4.685083648471835, "tokens_seen": 114360320 }, { "epoch": 0.03, "learning_rate": 0.00048754262788365093, "loss": 3.3925, "theoretical_loss": 4.684718549206607, "tokens_seen": 114425856 }, { "epoch": 0.03, "learning_rate": 0.00048753259779338017, "loss": 3.4082, "theoretical_loss": 4.6843537174991345, "tokens_seen": 114491392 }, { "epoch": 0.03, "learning_rate": 0.00048752256770310935, "loss": 3.3881, "theoretical_loss": 4.6839891530003595, "tokens_seen": 114556928 }, { "epoch": 0.03, "learning_rate": 0.00048751253761283853, "loss": 3.4909, "theoretical_loss": 4.683624855361876, "tokens_seen": 114622464 }, { "epoch": 0.03, "objective/train/docs_used": 76169, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.169177770614624, "objective/train/theoretical_loss": 4.68326082423593, "objective/train/tokens_used": 135148000, "theoretical_loss": 4.68326082423593, "tokens_seen": 114688000 }, { "epoch": 0.03, "learning_rate": 0.0004875025075225677, "loss": 3.1599, "theoretical_loss": 4.68326082423593, "tokens_seen": 114688000 }, { "epoch": 0.03, "learning_rate": 0.0004874924774322969, "loss": 3.744, "theoretical_loss": 4.682897059275422, "tokens_seen": 114753536 }, { "epoch": 0.03, "learning_rate": 0.00048748244734202607, "loss": 3.5137, "theoretical_loss": 4.682533560133901, "tokens_seen": 114819072 }, { "epoch": 0.03, "learning_rate": 0.0004874724172517553, "loss": 3.3442, "theoretical_loss": 4.682170326465565, "tokens_seen": 114884608 }, { "epoch": 0.03, "learning_rate": 0.00048746238716148443, "loss": 3.4037, "theoretical_loss": 4.681807357925257, "tokens_seen": 114950144 }, { "epoch": 0.03, "learning_rate": 0.00048745235707121367, "loss": 3.6075, "theoretical_loss": 4.681444654168468, "tokens_seen": 115015680 }, { "epoch": 0.03, "learning_rate": 0.0004874423269809428, "loss": 3.3067, "theoretical_loss": 4.68108221485133, "tokens_seen": 115081216 }, { "epoch": 0.03, "learning_rate": 0.00048743229689067203, "loss": 3.5086, "theoretical_loss": 4.680720039630617, "tokens_seen": 115146752 }, { "epoch": 0.03, "learning_rate": 0.0004874222668004012, "loss": 3.1917, "theoretical_loss": 4.680358128163747, "tokens_seen": 115212288 }, { "epoch": 0.03, "learning_rate": 0.0004874122367101304, "loss": 3.6048, "theoretical_loss": 4.679996480108773, "tokens_seen": 115277824 }, { "epoch": 0.03, "learning_rate": 0.0004874022066198596, "loss": 3.3722, "theoretical_loss": 4.6796350951243895, "tokens_seen": 115343360 }, { "epoch": 0.03, "learning_rate": 0.00048739217652958876, "loss": 3.491, "theoretical_loss": 4.679273972869922, "tokens_seen": 115408896 }, { "epoch": 0.03, "learning_rate": 0.00048738214643931794, "loss": 3.3156, "theoretical_loss": 4.678913113005333, "tokens_seen": 115474432 }, { "epoch": 0.04, "learning_rate": 0.00048737211634904717, "loss": 3.3652, "theoretical_loss": 4.6785525151912175, "tokens_seen": 115539968 }, { "epoch": 0.04, "learning_rate": 0.0004873620862587763, "loss": 3.3324, "theoretical_loss": 4.678192179088802, "tokens_seen": 115605504 }, { "epoch": 0.04, "learning_rate": 0.00048735205616850553, "loss": 3.6088, "theoretical_loss": 4.6778321043599425, "tokens_seen": 115671040 }, { "epoch": 0.04, "learning_rate": 0.0004873420260782347, "loss": 3.3234, "theoretical_loss": 4.677472290667122, "tokens_seen": 115736576 }, { "epoch": 0.04, "learning_rate": 0.0004873319959879639, "loss": 3.3235, "theoretical_loss": 4.677112737673453, "tokens_seen": 115802112 }, { "epoch": 0.04, "learning_rate": 0.0004873219658976931, "loss": 3.3194, "theoretical_loss": 4.676753445042669, "tokens_seen": 115867648 }, { "epoch": 0.04, "learning_rate": 0.00048731193580742226, "loss": 3.0035, "theoretical_loss": 4.676394412439132, "tokens_seen": 115933184 }, { "epoch": 0.04, "learning_rate": 0.00048730190571715144, "loss": 3.2693, "theoretical_loss": 4.6760356395278215, "tokens_seen": 115998720 }, { "epoch": 0.04, "learning_rate": 0.0004872918756268807, "loss": 3.3124, "theoretical_loss": 4.675677125974339, "tokens_seen": 116064256 }, { "epoch": 0.04, "learning_rate": 0.0004872818455366098, "loss": 3.4606, "theoretical_loss": 4.675318871444908, "tokens_seen": 116129792 }, { "epoch": 0.04, "learning_rate": 0.00048727181544633904, "loss": 3.3459, "theoretical_loss": 4.674960875606366, "tokens_seen": 116195328 }, { "epoch": 0.04, "learning_rate": 0.00048726178535606816, "loss": 3.463, "theoretical_loss": 4.674603138126168, "tokens_seen": 116260864 }, { "epoch": 0.04, "objective/train/docs_used": 76931, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.525822162628174, "objective/train/theoretical_loss": 4.674245658672382, "objective/train/tokens_used": 136786400, "theoretical_loss": 4.674245658672382, "tokens_seen": 116326400 }, { "epoch": 0.04, "learning_rate": 0.0004872517552657974, "loss": 3.5205, "theoretical_loss": 4.674245658672382, "tokens_seen": 116326400 }, { "epoch": 0.04, "learning_rate": 0.0004872417251755266, "loss": 3.2274, "theoretical_loss": 4.673888436913694, "tokens_seen": 116391936 }, { "epoch": 0.04, "learning_rate": 0.00048723169508525576, "loss": 3.524, "theoretical_loss": 4.673531472519397, "tokens_seen": 116457472 }, { "epoch": 0.04, "learning_rate": 0.000487221664994985, "loss": 3.5341, "theoretical_loss": 4.673174765159393, "tokens_seen": 116523008 }, { "epoch": 0.04, "learning_rate": 0.0004872116349047141, "loss": 3.3575, "theoretical_loss": 4.672818314504198, "tokens_seen": 116588544 }, { "epoch": 0.04, "learning_rate": 0.00048720160481444336, "loss": 3.2956, "theoretical_loss": 4.6724621202249335, "tokens_seen": 116654080 }, { "epoch": 0.04, "learning_rate": 0.00048719157472417254, "loss": 3.4414, "theoretical_loss": 4.672106181993324, "tokens_seen": 116719616 }, { "epoch": 0.04, "learning_rate": 0.0004871815446339017, "loss": 3.4029, "theoretical_loss": 4.6717504994817, "tokens_seen": 116785152 }, { "epoch": 0.04, "learning_rate": 0.0004871715145436309, "loss": 3.3539, "theoretical_loss": 4.671395072362996, "tokens_seen": 116850688 }, { "epoch": 0.04, "learning_rate": 0.00048716148445336014, "loss": 3.7625, "theoretical_loss": 4.671039900310747, "tokens_seen": 116916224 }, { "epoch": 0.04, "learning_rate": 0.00048715145436308927, "loss": 3.2591, "theoretical_loss": 4.670684982999088, "tokens_seen": 116981760 }, { "epoch": 0.04, "learning_rate": 0.0004871414242728185, "loss": 3.4151, "theoretical_loss": 4.670330320102753, "tokens_seen": 117047296 }, { "epoch": 0.04, "learning_rate": 0.00048713139418254763, "loss": 3.4562, "theoretical_loss": 4.669975911297072, "tokens_seen": 117112832 }, { "epoch": 0.04, "learning_rate": 0.00048712136409227686, "loss": 3.4337, "theoretical_loss": 4.669621756257971, "tokens_seen": 117178368 }, { "epoch": 0.04, "learning_rate": 0.00048711133400200604, "loss": 3.3579, "theoretical_loss": 4.669267854661973, "tokens_seen": 117243904 }, { "epoch": 0.04, "learning_rate": 0.0004871013039117352, "loss": 3.3642, "theoretical_loss": 4.668914206186189, "tokens_seen": 117309440 }, { "epoch": 0.04, "learning_rate": 0.0004870912738214644, "loss": 3.4332, "theoretical_loss": 4.6685608105083265, "tokens_seen": 117374976 }, { "epoch": 0.04, "learning_rate": 0.0004870812437311936, "loss": 3.5261, "theoretical_loss": 4.66820766730668, "tokens_seen": 117440512 }, { "epoch": 0.04, "learning_rate": 0.00048707121364092277, "loss": 3.2805, "theoretical_loss": 4.667854776260132, "tokens_seen": 117506048 }, { "epoch": 0.04, "learning_rate": 0.000487061183550652, "loss": 3.5186, "theoretical_loss": 4.667502137048155, "tokens_seen": 117571584 }, { "epoch": 0.04, "learning_rate": 0.00048705115346038113, "loss": 3.3778, "theoretical_loss": 4.667149749350805, "tokens_seen": 117637120 }, { "epoch": 0.04, "learning_rate": 0.00048704112337011037, "loss": 3.1395, "theoretical_loss": 4.666797612848723, "tokens_seen": 117702656 }, { "epoch": 0.04, "learning_rate": 0.00048703109327983955, "loss": 3.3278, "theoretical_loss": 4.666445727223134, "tokens_seen": 117768192 }, { "epoch": 0.04, "learning_rate": 0.00048702106318956873, "loss": 3.2112, "theoretical_loss": 4.666094092155843, "tokens_seen": 117833728 }, { "epoch": 0.04, "learning_rate": 0.0004870110330992979, "loss": 3.6812, "theoretical_loss": 4.665742707329238, "tokens_seen": 117899264 }, { "epoch": 0.04, "objective/train/docs_used": 78354, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.753464460372925, "objective/train/theoretical_loss": 4.665391572426282, "objective/train/tokens_used": 138424800, "theoretical_loss": 4.665391572426282, "tokens_seen": 117964800 }, { "epoch": 0.04, "learning_rate": 0.0004870010030090271, "loss": 3.4575, "theoretical_loss": 4.665391572426282, "tokens_seen": 117964800 }, { "epoch": 0.04, "learning_rate": 0.00048699097291875627, "loss": 3.2806, "theoretical_loss": 4.665040687130518, "tokens_seen": 118030336 }, { "epoch": 0.04, "learning_rate": 0.0004869809428284855, "loss": 3.2856, "theoretical_loss": 4.664690051126065, "tokens_seen": 118095872 }, { "epoch": 0.04, "learning_rate": 0.00048697091273821463, "loss": 3.6047, "theoretical_loss": 4.664339664097617, "tokens_seen": 118161408 }, { "epoch": 0.04, "learning_rate": 0.00048696088264794387, "loss": 3.3602, "theoretical_loss": 4.66398952573044, "tokens_seen": 118226944 }, { "epoch": 0.04, "learning_rate": 0.000486950852557673, "loss": 3.6397, "theoretical_loss": 4.663639635710373, "tokens_seen": 118292480 }, { "epoch": 0.04, "learning_rate": 0.00048694082246740223, "loss": 3.5041, "theoretical_loss": 4.663289993723826, "tokens_seen": 118358016 }, { "epoch": 0.04, "learning_rate": 0.0004869307923771314, "loss": 3.6349, "theoretical_loss": 4.662940599457777, "tokens_seen": 118423552 }, { "epoch": 0.04, "learning_rate": 0.0004869207622868606, "loss": 3.311, "theoretical_loss": 4.662591452599774, "tokens_seen": 118489088 }, { "epoch": 0.04, "learning_rate": 0.0004869107321965898, "loss": 3.3616, "theoretical_loss": 4.662242552837929, "tokens_seen": 118554624 }, { "epoch": 0.04, "learning_rate": 0.00048690070210631896, "loss": 3.5774, "theoretical_loss": 4.661893899860923, "tokens_seen": 118620160 }, { "epoch": 0.04, "learning_rate": 0.00048689067201604814, "loss": 3.5329, "theoretical_loss": 4.6615454933579965, "tokens_seen": 118685696 }, { "epoch": 0.04, "learning_rate": 0.00048688064192577737, "loss": 3.3496, "theoretical_loss": 4.661197333018957, "tokens_seen": 118751232 }, { "epoch": 0.04, "learning_rate": 0.0004868706118355065, "loss": 3.1691, "theoretical_loss": 4.66084941853417, "tokens_seen": 118816768 }, { "epoch": 0.04, "learning_rate": 0.00048686058174523573, "loss": 3.4726, "theoretical_loss": 4.6605017495945615, "tokens_seen": 118882304 }, { "epoch": 0.04, "learning_rate": 0.0004868505516549649, "loss": 3.2301, "theoretical_loss": 4.660154325891618, "tokens_seen": 118947840 }, { "epoch": 0.04, "learning_rate": 0.0004868405215646941, "loss": 3.3908, "theoretical_loss": 4.659807147117382, "tokens_seen": 119013376 }, { "epoch": 0.04, "learning_rate": 0.0004868304914744233, "loss": 3.5186, "theoretical_loss": 4.6594602129644525, "tokens_seen": 119078912 }, { "epoch": 0.04, "learning_rate": 0.00048682046138415246, "loss": 3.5771, "theoretical_loss": 4.659113523125981, "tokens_seen": 119144448 }, { "epoch": 0.04, "learning_rate": 0.00048681043129388164, "loss": 3.4647, "theoretical_loss": 4.6587670772956775, "tokens_seen": 119209984 }, { "epoch": 0.04, "learning_rate": 0.0004868004012036109, "loss": 3.359, "theoretical_loss": 4.658420875167799, "tokens_seen": 119275520 }, { "epoch": 0.04, "learning_rate": 0.00048679037111334, "loss": 3.2057, "theoretical_loss": 4.658074916437155, "tokens_seen": 119341056 }, { "epoch": 0.04, "learning_rate": 0.00048678034102306924, "loss": 3.5888, "theoretical_loss": 4.657729200799105, "tokens_seen": 119406592 }, { "epoch": 0.04, "learning_rate": 0.00048677031093279836, "loss": 3.3121, "theoretical_loss": 4.657383727949558, "tokens_seen": 119472128 }, { "epoch": 0.04, "learning_rate": 0.0004867602808425276, "loss": 3.4127, "theoretical_loss": 4.657038497584967, "tokens_seen": 119537664 }, { "epoch": 0.04, "objective/train/docs_used": 78905, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3278000354766846, "objective/train/theoretical_loss": 4.656693509402331, "objective/train/tokens_used": 140063200, "theoretical_loss": 4.656693509402331, "tokens_seen": 119603200 }, { "epoch": 0.04, "learning_rate": 0.0004867502507522568, "loss": 3.5357, "theoretical_loss": 4.656693509402331, "tokens_seen": 119603200 }, { "epoch": 0.04, "learning_rate": 0.00048674022066198596, "loss": 3.3471, "theoretical_loss": 4.6563487630991975, "tokens_seen": 119668736 }, { "epoch": 0.04, "learning_rate": 0.00048673019057171514, "loss": 3.5287, "theoretical_loss": 4.656004258373651, "tokens_seen": 119734272 }, { "epoch": 0.04, "learning_rate": 0.0004867201604814443, "loss": 3.5142, "theoretical_loss": 4.655659994924323, "tokens_seen": 119799808 }, { "epoch": 0.04, "learning_rate": 0.0004867101303911735, "loss": 3.2315, "theoretical_loss": 4.655315972450383, "tokens_seen": 119865344 }, { "epoch": 0.04, "learning_rate": 0.00048670010030090274, "loss": 3.1649, "theoretical_loss": 4.65497219065154, "tokens_seen": 119930880 }, { "epoch": 0.04, "learning_rate": 0.00048669007021063187, "loss": 3.5037, "theoretical_loss": 4.654628649228041, "tokens_seen": 119996416 }, { "epoch": 0.04, "learning_rate": 0.0004866800401203611, "loss": 3.2458, "theoretical_loss": 4.654285347880672, "tokens_seen": 120061952 }, { "epoch": 0.04, "learning_rate": 0.0004866700100300903, "loss": 3.407, "theoretical_loss": 4.653942286310749, "tokens_seen": 120127488 }, { "epoch": 0.04, "learning_rate": 0.00048665997993981947, "loss": 3.4908, "theoretical_loss": 4.653599464220129, "tokens_seen": 120193024 }, { "epoch": 0.04, "learning_rate": 0.00048664994984954865, "loss": 3.3596, "theoretical_loss": 4.653256881311198, "tokens_seen": 120258560 }, { "epoch": 0.04, "learning_rate": 0.00048663991975927783, "loss": 3.489, "theoretical_loss": 4.6529145372868745, "tokens_seen": 120324096 }, { "epoch": 0.04, "learning_rate": 0.000486629889669007, "loss": 3.3819, "theoretical_loss": 4.652572431850608, "tokens_seen": 120389632 }, { "epoch": 0.04, "learning_rate": 0.00048661985957873624, "loss": 3.2704, "theoretical_loss": 4.652230564706377, "tokens_seen": 120455168 }, { "epoch": 0.04, "learning_rate": 0.00048660982948846537, "loss": 3.2807, "theoretical_loss": 4.651888935558688, "tokens_seen": 120520704 }, { "epoch": 0.04, "learning_rate": 0.0004865997993981946, "loss": 3.4143, "theoretical_loss": 4.651547544112575, "tokens_seen": 120586240 }, { "epoch": 0.04, "learning_rate": 0.00048658976930792373, "loss": 3.3822, "theoretical_loss": 4.651206390073597, "tokens_seen": 120651776 }, { "epoch": 0.04, "learning_rate": 0.00048657973921765297, "loss": 3.0172, "theoretical_loss": 4.650865473147837, "tokens_seen": 120717312 }, { "epoch": 0.04, "learning_rate": 0.00048656970912738215, "loss": 3.2458, "theoretical_loss": 4.650524793041903, "tokens_seen": 120782848 }, { "epoch": 0.04, "learning_rate": 0.00048655967903711133, "loss": 3.3165, "theoretical_loss": 4.650184349462922, "tokens_seen": 120848384 }, { "epoch": 0.04, "learning_rate": 0.0004865496489468405, "loss": 3.5201, "theoretical_loss": 4.649844142118544, "tokens_seen": 120913920 }, { "epoch": 0.04, "learning_rate": 0.00048653961885656975, "loss": 3.2466, "theoretical_loss": 4.6495041707169396, "tokens_seen": 120979456 }, { "epoch": 0.04, "learning_rate": 0.0004865295887662989, "loss": 3.4625, "theoretical_loss": 4.649164434966794, "tokens_seen": 121044992 }, { "epoch": 0.04, "learning_rate": 0.0004865195586760281, "loss": 3.4472, "theoretical_loss": 4.648824934577313, "tokens_seen": 121110528 }, { "epoch": 0.04, "learning_rate": 0.00048650952858575724, "loss": 3.454, "theoretical_loss": 4.648485669258216, "tokens_seen": 121176064 }, { "epoch": 0.04, "objective/train/docs_used": 79692, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7176690101623535, "objective/train/theoretical_loss": 4.648146638719739, "objective/train/tokens_used": 141701600, "theoretical_loss": 4.648146638719739, "tokens_seen": 121241600 }, { "epoch": 0.04, "learning_rate": 0.00048649949849548647, "loss": 3.6455, "theoretical_loss": 4.648146638719739, "tokens_seen": 121241600 }, { "epoch": 0.04, "learning_rate": 0.00048648946840521565, "loss": 3.2591, "theoretical_loss": 4.647807842672631, "tokens_seen": 121307136 }, { "epoch": 0.04, "learning_rate": 0.00048647943831494483, "loss": 3.1074, "theoretical_loss": 4.647469280828153, "tokens_seen": 121372672 }, { "epoch": 0.04, "learning_rate": 0.00048646940822467407, "loss": 3.625, "theoretical_loss": 4.647130952898077, "tokens_seen": 121438208 }, { "epoch": 0.04, "learning_rate": 0.0004864593781344032, "loss": 3.6184, "theoretical_loss": 4.646792858594686, "tokens_seen": 121503744 }, { "epoch": 0.04, "learning_rate": 0.00048644934804413243, "loss": 3.3147, "theoretical_loss": 4.64645499763077, "tokens_seen": 121569280 }, { "epoch": 0.04, "learning_rate": 0.0004864393179538616, "loss": 3.3483, "theoretical_loss": 4.646117369719629, "tokens_seen": 121634816 }, { "epoch": 0.04, "learning_rate": 0.0004864292878635908, "loss": 3.6571, "theoretical_loss": 4.645779974575069, "tokens_seen": 121700352 }, { "epoch": 0.04, "learning_rate": 0.00048641925777332, "loss": 3.392, "theoretical_loss": 4.6454428119113995, "tokens_seen": 121765888 }, { "epoch": 0.04, "learning_rate": 0.00048640922768304916, "loss": 3.4246, "theoretical_loss": 4.6451058814434365, "tokens_seen": 121831424 }, { "epoch": 0.04, "learning_rate": 0.00048639919759277834, "loss": 3.2547, "theoretical_loss": 4.644769182886495, "tokens_seen": 121896960 }, { "epoch": 0.04, "learning_rate": 0.00048638916750250757, "loss": 3.4761, "theoretical_loss": 4.644432715956399, "tokens_seen": 121962496 }, { "epoch": 0.04, "learning_rate": 0.0004863791374122367, "loss": 3.3089, "theoretical_loss": 4.644096480369466, "tokens_seen": 122028032 }, { "epoch": 0.04, "learning_rate": 0.00048636910732196593, "loss": 3.4689, "theoretical_loss": 4.643760475842518, "tokens_seen": 122093568 }, { "epoch": 0.04, "learning_rate": 0.0004863590772316951, "loss": 3.3779, "theoretical_loss": 4.6434247020928705, "tokens_seen": 122159104 }, { "epoch": 0.04, "learning_rate": 0.0004863490471414243, "loss": 3.3198, "theoretical_loss": 4.643089158838341, "tokens_seen": 122224640 }, { "epoch": 0.04, "learning_rate": 0.0004863390170511535, "loss": 3.4073, "theoretical_loss": 4.642753845797243, "tokens_seen": 122290176 }, { "epoch": 0.04, "learning_rate": 0.00048632898696088266, "loss": 3.3571, "theoretical_loss": 4.642418762688379, "tokens_seen": 122355712 }, { "epoch": 0.04, "learning_rate": 0.00048631895687061184, "loss": 3.4689, "theoretical_loss": 4.642083909231053, "tokens_seen": 122421248 }, { "epoch": 0.04, "learning_rate": 0.0004863089267803411, "loss": 3.1746, "theoretical_loss": 4.641749285145057, "tokens_seen": 122486784 }, { "epoch": 0.04, "learning_rate": 0.0004862988966900702, "loss": 3.339, "theoretical_loss": 4.641414890150675, "tokens_seen": 122552320 }, { "epoch": 0.04, "learning_rate": 0.00048628886659979944, "loss": 3.2195, "theoretical_loss": 4.641080723968684, "tokens_seen": 122617856 }, { "epoch": 0.04, "learning_rate": 0.00048627883650952857, "loss": 3.3498, "theoretical_loss": 4.6407467863203475, "tokens_seen": 122683392 }, { "epoch": 0.04, "learning_rate": 0.0004862688064192578, "loss": 3.6232, "theoretical_loss": 4.640413076927418, "tokens_seen": 122748928 }, { "epoch": 0.04, "learning_rate": 0.000486258776328987, "loss": 3.3464, "theoretical_loss": 4.6400795955121374, "tokens_seen": 122814464 }, { "epoch": 0.04, "objective/train/docs_used": 81051, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.561691999435425, "objective/train/theoretical_loss": 4.639746341797229, "objective/train/tokens_used": 143340000, "theoretical_loss": 4.639746341797229, "tokens_seen": 122880000 }, { "epoch": 0.04, "learning_rate": 0.00048624874623871616, "loss": 3.3018, "theoretical_loss": 4.639746341797229, "tokens_seen": 122880000 }, { "epoch": 0.04, "learning_rate": 0.00048623871614844534, "loss": 3.3855, "theoretical_loss": 4.639413315505905, "tokens_seen": 122945536 }, { "epoch": 0.04, "learning_rate": 0.0004862286860581745, "loss": 3.3287, "theoretical_loss": 4.639080516361861, "tokens_seen": 123011072 }, { "epoch": 0.04, "learning_rate": 0.0004862186559679037, "loss": 3.3515, "theoretical_loss": 4.638747944089273, "tokens_seen": 123076608 }, { "epoch": 0.04, "learning_rate": 0.00048620862587763294, "loss": 3.5609, "theoretical_loss": 4.638415598412799, "tokens_seen": 123142144 }, { "epoch": 0.04, "learning_rate": 0.00048619859578736207, "loss": 3.1183, "theoretical_loss": 4.638083479057579, "tokens_seen": 123207680 }, { "epoch": 0.04, "learning_rate": 0.0004861885656970913, "loss": 3.2753, "theoretical_loss": 4.637751585749234, "tokens_seen": 123273216 }, { "epoch": 0.04, "learning_rate": 0.0004861785356068205, "loss": 3.3623, "theoretical_loss": 4.6374199182138565, "tokens_seen": 123338752 }, { "epoch": 0.04, "learning_rate": 0.00048616850551654967, "loss": 3.2958, "theoretical_loss": 4.637088476178025, "tokens_seen": 123404288 }, { "epoch": 0.04, "learning_rate": 0.00048615847542627885, "loss": 3.3467, "theoretical_loss": 4.636757259368787, "tokens_seen": 123469824 }, { "epoch": 0.04, "learning_rate": 0.00048614844533600803, "loss": 3.6196, "theoretical_loss": 4.636426267513668, "tokens_seen": 123535360 }, { "epoch": 0.04, "learning_rate": 0.0004861384152457372, "loss": 3.4397, "theoretical_loss": 4.636095500340669, "tokens_seen": 123600896 }, { "epoch": 0.04, "learning_rate": 0.00048612838515546644, "loss": 3.3183, "theoretical_loss": 4.635764957578261, "tokens_seen": 123666432 }, { "epoch": 0.04, "learning_rate": 0.00048611835506519557, "loss": 3.5385, "theoretical_loss": 4.635434638955388, "tokens_seen": 123731968 }, { "epoch": 0.04, "learning_rate": 0.0004861083249749248, "loss": 3.4339, "theoretical_loss": 4.635104544201465, "tokens_seen": 123797504 }, { "epoch": 0.04, "learning_rate": 0.00048609829488465393, "loss": 3.4177, "theoretical_loss": 4.634774673046376, "tokens_seen": 123863040 }, { "epoch": 0.04, "learning_rate": 0.00048608826479438317, "loss": 3.5387, "theoretical_loss": 4.634445025220475, "tokens_seen": 123928576 }, { "epoch": 0.04, "learning_rate": 0.00048607823470411235, "loss": 3.2727, "theoretical_loss": 4.634115600454582, "tokens_seen": 123994112 }, { "epoch": 0.04, "learning_rate": 0.00048606820461384153, "loss": 3.5262, "theoretical_loss": 4.633786398479983, "tokens_seen": 124059648 }, { "epoch": 0.04, "learning_rate": 0.0004860581745235707, "loss": 3.3889, "theoretical_loss": 4.6334574190284314, "tokens_seen": 124125184 }, { "epoch": 0.04, "learning_rate": 0.00048604814443329995, "loss": 3.1077, "theoretical_loss": 4.633128661832145, "tokens_seen": 124190720 }, { "epoch": 0.04, "learning_rate": 0.0004860381143430291, "loss": 3.5107, "theoretical_loss": 4.632800126623803, "tokens_seen": 124256256 }, { "epoch": 0.04, "learning_rate": 0.0004860280842527583, "loss": 3.4446, "theoretical_loss": 4.632471813136547, "tokens_seen": 124321792 }, { "epoch": 0.04, "learning_rate": 0.00048601805416248744, "loss": 3.3011, "theoretical_loss": 4.632143721103983, "tokens_seen": 124387328 }, { "epoch": 0.04, "learning_rate": 0.00048600802407221667, "loss": 3.4728, "theoretical_loss": 4.631815850260173, "tokens_seen": 124452864 }, { "epoch": 0.04, "objective/train/docs_used": 81846, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6328256130218506, "objective/train/theoretical_loss": 4.631488200339643, "objective/train/tokens_used": 144978400, "theoretical_loss": 4.631488200339643, "tokens_seen": 124518400 }, { "epoch": 0.04, "learning_rate": 0.00048599799398194585, "loss": 3.3451, "theoretical_loss": 4.631488200339643, "tokens_seen": 124518400 }, { "epoch": 0.04, "learning_rate": 0.00048598796389167503, "loss": 3.3085, "theoretical_loss": 4.63116077107737, "tokens_seen": 124583936 }, { "epoch": 0.04, "learning_rate": 0.0004859779338014042, "loss": 3.3846, "theoretical_loss": 4.630833562208797, "tokens_seen": 124649472 }, { "epoch": 0.04, "learning_rate": 0.0004859679037111334, "loss": 3.4496, "theoretical_loss": 4.630506573469815, "tokens_seen": 124715008 }, { "epoch": 0.04, "learning_rate": 0.0004859578736208626, "loss": 3.0359, "theoretical_loss": 4.630179804596775, "tokens_seen": 124780544 }, { "epoch": 0.04, "learning_rate": 0.0004859478435305918, "loss": 3.3915, "theoretical_loss": 4.629853255326481, "tokens_seen": 124846080 }, { "epoch": 0.04, "learning_rate": 0.00048593781344032094, "loss": 3.1342, "theoretical_loss": 4.629526925396189, "tokens_seen": 124911616 }, { "epoch": 0.04, "learning_rate": 0.0004859277833500502, "loss": 3.5865, "theoretical_loss": 4.6292008145436085, "tokens_seen": 124977152 }, { "epoch": 0.04, "learning_rate": 0.0004859177532597793, "loss": 3.3983, "theoretical_loss": 4.628874922506897, "tokens_seen": 125042688 }, { "epoch": 0.04, "learning_rate": 0.00048590772316950854, "loss": 3.3449, "theoretical_loss": 4.628549249024666, "tokens_seen": 125108224 }, { "epoch": 0.04, "learning_rate": 0.0004858976930792377, "loss": 3.3839, "theoretical_loss": 4.628223793835975, "tokens_seen": 125173760 }, { "epoch": 0.04, "learning_rate": 0.0004858876629889669, "loss": 3.1372, "theoretical_loss": 4.627898556680327, "tokens_seen": 125239296 }, { "epoch": 0.04, "learning_rate": 0.0004858776328986961, "loss": 3.2933, "theoretical_loss": 4.627573537297678, "tokens_seen": 125304832 }, { "epoch": 0.04, "learning_rate": 0.0004858676028084253, "loss": 3.506, "theoretical_loss": 4.627248735428427, "tokens_seen": 125370368 }, { "epoch": 0.04, "learning_rate": 0.00048585757271815444, "loss": 3.4041, "theoretical_loss": 4.6269241508134185, "tokens_seen": 125435904 }, { "epoch": 0.04, "learning_rate": 0.0004858475426278837, "loss": 3.5242, "theoretical_loss": 4.6265997831939405, "tokens_seen": 125501440 }, { "epoch": 0.04, "learning_rate": 0.0004858375125376128, "loss": 3.3199, "theoretical_loss": 4.6262756323117245, "tokens_seen": 125566976 }, { "epoch": 0.04, "learning_rate": 0.00048582748244734204, "loss": 3.1673, "theoretical_loss": 4.625951697908944, "tokens_seen": 125632512 }, { "epoch": 0.04, "learning_rate": 0.0004858174523570712, "loss": 3.5243, "theoretical_loss": 4.625627979728212, "tokens_seen": 125698048 }, { "epoch": 0.04, "learning_rate": 0.0004858074222668004, "loss": 3.404, "theoretical_loss": 4.625304477512584, "tokens_seen": 125763584 }, { "epoch": 0.04, "learning_rate": 0.0004857973921765296, "loss": 3.2105, "theoretical_loss": 4.624981191005554, "tokens_seen": 125829120 }, { "epoch": 0.04, "learning_rate": 0.00048578736208625877, "loss": 3.0283, "theoretical_loss": 4.624658119951052, "tokens_seen": 125894656 }, { "epoch": 0.04, "learning_rate": 0.00048577733199598795, "loss": 3.1752, "theoretical_loss": 4.624335264093447, "tokens_seen": 125960192 }, { "epoch": 0.04, "learning_rate": 0.0004857673019057172, "loss": 3.5489, "theoretical_loss": 4.624012623177544, "tokens_seen": 126025728 }, { "epoch": 0.04, "learning_rate": 0.0004857572718154463, "loss": 3.3505, "theoretical_loss": 4.623690196948582, "tokens_seen": 126091264 }, { "epoch": 0.04, "objective/train/docs_used": 83055, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7693979740142822, "objective/train/theoretical_loss": 4.623367985152234, "objective/train/tokens_used": 146616800, "theoretical_loss": 4.623367985152234, "tokens_seen": 126156800 }, { "epoch": 0.04, "learning_rate": 0.00048574724172517554, "loss": 3.554, "theoretical_loss": 4.623367985152234, "tokens_seen": 126156800 }, { "epoch": 0.04, "learning_rate": 0.00048573721163490467, "loss": 3.4829, "theoretical_loss": 4.623045987534609, "tokens_seen": 126222336 }, { "epoch": 0.04, "learning_rate": 0.0004857271815446339, "loss": 3.5869, "theoretical_loss": 4.622724203842246, "tokens_seen": 126287872 }, { "epoch": 0.04, "learning_rate": 0.00048571715145436314, "loss": 3.3389, "theoretical_loss": 4.622402633822114, "tokens_seen": 126353408 }, { "epoch": 0.04, "learning_rate": 0.00048570712136409227, "loss": 3.3537, "theoretical_loss": 4.622081277221616, "tokens_seen": 126418944 }, { "epoch": 0.04, "learning_rate": 0.0004856970912738215, "loss": 3.2713, "theoretical_loss": 4.62176013378858, "tokens_seen": 126484480 }, { "epoch": 0.04, "learning_rate": 0.0004856870611835507, "loss": 3.4149, "theoretical_loss": 4.621439203271267, "tokens_seen": 126550016 }, { "epoch": 0.04, "learning_rate": 0.00048567703109327987, "loss": 3.4411, "theoretical_loss": 4.621118485418362, "tokens_seen": 126615552 }, { "epoch": 0.04, "learning_rate": 0.00048566700100300905, "loss": 3.5645, "theoretical_loss": 4.620797979978978, "tokens_seen": 126681088 }, { "epoch": 0.04, "learning_rate": 0.00048565697091273823, "loss": 3.4029, "theoretical_loss": 4.620477686702651, "tokens_seen": 126746624 }, { "epoch": 0.04, "learning_rate": 0.0004856469408224674, "loss": 3.5134, "theoretical_loss": 4.620157605339347, "tokens_seen": 126812160 }, { "epoch": 0.04, "learning_rate": 0.00048563691073219664, "loss": 3.2403, "theoretical_loss": 4.619837735639452, "tokens_seen": 126877696 }, { "epoch": 0.04, "learning_rate": 0.00048562688064192577, "loss": 3.2403, "theoretical_loss": 4.619518077353776, "tokens_seen": 126943232 }, { "epoch": 0.04, "learning_rate": 0.000485616850551655, "loss": 3.3605, "theoretical_loss": 4.619198630233547, "tokens_seen": 127008768 }, { "epoch": 0.04, "learning_rate": 0.00048560682046138413, "loss": 3.6782, "theoretical_loss": 4.6188793940304205, "tokens_seen": 127074304 }, { "epoch": 0.04, "learning_rate": 0.00048559679037111337, "loss": 3.291, "theoretical_loss": 4.618560368496466, "tokens_seen": 127139840 }, { "epoch": 0.04, "learning_rate": 0.00048558676028084255, "loss": 3.2539, "theoretical_loss": 4.618241553384175, "tokens_seen": 127205376 }, { "epoch": 0.04, "learning_rate": 0.00048557673019057173, "loss": 3.3349, "theoretical_loss": 4.617922948446459, "tokens_seen": 127270912 }, { "epoch": 0.04, "learning_rate": 0.0004855667001003009, "loss": 3.3863, "theoretical_loss": 4.617604553436642, "tokens_seen": 127336448 }, { "epoch": 0.04, "learning_rate": 0.00048555667001003015, "loss": 3.3637, "theoretical_loss": 4.617286368108466, "tokens_seen": 127401984 }, { "epoch": 0.04, "learning_rate": 0.0004855466399197593, "loss": 3.2942, "theoretical_loss": 4.6169683922160925, "tokens_seen": 127467520 }, { "epoch": 0.04, "learning_rate": 0.0004855366098294885, "loss": 3.4269, "theoretical_loss": 4.616650625514091, "tokens_seen": 127533056 }, { "epoch": 0.04, "learning_rate": 0.00048552657973921764, "loss": 3.3937, "theoretical_loss": 4.616333067757449, "tokens_seen": 127598592 }, { "epoch": 0.04, "learning_rate": 0.00048551654964894687, "loss": 3.4983, "theoretical_loss": 4.616015718701563, "tokens_seen": 127664128 }, { "epoch": 0.04, "learning_rate": 0.00048550651955867605, "loss": 3.3911, "theoretical_loss": 4.615698578102245, "tokens_seen": 127729664 }, { "epoch": 0.04, "objective/train/docs_used": 83592, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.880964994430542, "objective/train/theoretical_loss": 4.615381645715717, "objective/train/tokens_used": 148255200, "theoretical_loss": 4.615381645715717, "tokens_seen": 127795200 }, { "epoch": 0.04, "learning_rate": 0.00048549648946840523, "loss": 3.5533, "theoretical_loss": 4.615381645715717, "tokens_seen": 127795200 }, { "epoch": 0.04, "learning_rate": 0.0004854864593781344, "loss": 3.4536, "theoretical_loss": 4.615064921298608, "tokens_seen": 127860736 }, { "epoch": 0.04, "learning_rate": 0.0004854764292878636, "loss": 3.377, "theoretical_loss": 4.61474840460796, "tokens_seen": 127926272 }, { "epoch": 0.04, "learning_rate": 0.0004854663991975928, "loss": 3.5236, "theoretical_loss": 4.614432095401219, "tokens_seen": 127991808 }, { "epoch": 0.04, "learning_rate": 0.000485456369107322, "loss": 3.4254, "theoretical_loss": 4.614115993436242, "tokens_seen": 128057344 }, { "epoch": 0.04, "learning_rate": 0.00048544633901705114, "loss": 3.4557, "theoretical_loss": 4.613800098471291, "tokens_seen": 128122880 }, { "epoch": 0.04, "learning_rate": 0.0004854363089267804, "loss": 3.5225, "theoretical_loss": 4.613484410265032, "tokens_seen": 128188416 }, { "epoch": 0.04, "learning_rate": 0.0004854262788365095, "loss": 3.3142, "theoretical_loss": 4.613168928576538, "tokens_seen": 128253952 }, { "epoch": 0.04, "learning_rate": 0.00048541624874623874, "loss": 3.3601, "theoretical_loss": 4.612853653165283, "tokens_seen": 128319488 }, { "epoch": 0.04, "learning_rate": 0.0004854062186559679, "loss": 3.4458, "theoretical_loss": 4.612538583791146, "tokens_seen": 128385024 }, { "epoch": 0.04, "learning_rate": 0.0004853961885656971, "loss": 3.1733, "theoretical_loss": 4.612223720214407, "tokens_seen": 128450560 }, { "epoch": 0.04, "learning_rate": 0.0004853861584754263, "loss": 3.2368, "theoretical_loss": 4.611909062195749, "tokens_seen": 128516096 }, { "epoch": 0.04, "learning_rate": 0.0004853761283851555, "loss": 3.3807, "theoretical_loss": 4.61159460949625, "tokens_seen": 128581632 }, { "epoch": 0.04, "learning_rate": 0.00048536609829488464, "loss": 3.3186, "theoretical_loss": 4.611280361877393, "tokens_seen": 128647168 }, { "epoch": 0.04, "learning_rate": 0.0004853560682046139, "loss": 3.4312, "theoretical_loss": 4.610966319101056, "tokens_seen": 128712704 }, { "epoch": 0.04, "learning_rate": 0.000485346038114343, "loss": 3.3996, "theoretical_loss": 4.610652480929515, "tokens_seen": 128778240 }, { "epoch": 0.04, "learning_rate": 0.00048533600802407224, "loss": 3.3225, "theoretical_loss": 4.610338847125445, "tokens_seen": 128843776 }, { "epoch": 0.04, "learning_rate": 0.0004853259779338014, "loss": 3.5473, "theoretical_loss": 4.610025417451913, "tokens_seen": 128909312 }, { "epoch": 0.04, "learning_rate": 0.0004853159478435306, "loss": 3.339, "theoretical_loss": 4.6097121916723856, "tokens_seen": 128974848 }, { "epoch": 0.04, "learning_rate": 0.0004853059177532598, "loss": 3.5529, "theoretical_loss": 4.609399169550718, "tokens_seen": 129040384 }, { "epoch": 0.04, "learning_rate": 0.00048529588766298897, "loss": 3.282, "theoretical_loss": 4.609086350851165, "tokens_seen": 129105920 }, { "epoch": 0.04, "learning_rate": 0.00048528585757271815, "loss": 3.3268, "theoretical_loss": 4.6087737353383655, "tokens_seen": 129171456 }, { "epoch": 0.04, "learning_rate": 0.0004852758274824474, "loss": 3.2374, "theoretical_loss": 4.6084613227773605, "tokens_seen": 129236992 }, { "epoch": 0.04, "learning_rate": 0.0004852657973921765, "loss": 3.2591, "theoretical_loss": 4.608149112933571, "tokens_seen": 129302528 }, { "epoch": 0.04, "learning_rate": 0.00048525576730190574, "loss": 3.0899, "theoretical_loss": 4.607837105572816, "tokens_seen": 129368064 }, { "epoch": 0.04, "objective/train/docs_used": 84814, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1867215633392334, "objective/train/theoretical_loss": 4.607525300461299, "objective/train/tokens_used": 149893600, "theoretical_loss": 4.607525300461299, "tokens_seen": 129433600 }, { "epoch": 0.04, "learning_rate": 0.00048524573721163487, "loss": 3.1172, "theoretical_loss": 4.607525300461299, "tokens_seen": 129433600 }, { "epoch": 0.04, "learning_rate": 0.0004852357071213641, "loss": 3.1282, "theoretical_loss": 4.607213697365613, "tokens_seen": 129499136 }, { "epoch": 0.04, "learning_rate": 0.0004852256770310933, "loss": 3.3892, "theoretical_loss": 4.606902296052739, "tokens_seen": 129564672 }, { "epoch": 0.04, "learning_rate": 0.00048521564694082247, "loss": 3.2205, "theoretical_loss": 4.6065910962900425, "tokens_seen": 129630208 }, { "epoch": 0.04, "learning_rate": 0.00048520561685055165, "loss": 3.4858, "theoretical_loss": 4.606280097845277, "tokens_seen": 129695744 }, { "epoch": 0.04, "learning_rate": 0.0004851955867602809, "loss": 3.2573, "theoretical_loss": 4.60596930048658, "tokens_seen": 129761280 }, { "epoch": 0.04, "learning_rate": 0.00048518555667001, "loss": 3.3325, "theoretical_loss": 4.605658703982471, "tokens_seen": 129826816 }, { "epoch": 0.04, "learning_rate": 0.00048517552657973925, "loss": 3.3328, "theoretical_loss": 4.6053483081018545, "tokens_seen": 129892352 }, { "epoch": 0.04, "learning_rate": 0.0004851654964894684, "loss": 3.3861, "theoretical_loss": 4.605038112614018, "tokens_seen": 129957888 }, { "epoch": 0.04, "learning_rate": 0.0004851554663991976, "loss": 3.1793, "theoretical_loss": 4.604728117288631, "tokens_seen": 130023424 }, { "epoch": 0.04, "learning_rate": 0.0004851454363089268, "loss": 3.275, "theoretical_loss": 4.604418321895739, "tokens_seen": 130088960 }, { "epoch": 0.04, "learning_rate": 0.00048513540621865597, "loss": 3.4315, "theoretical_loss": 4.604108726205774, "tokens_seen": 130154496 }, { "epoch": 0.04, "learning_rate": 0.00048512537612838515, "loss": 3.3219, "theoretical_loss": 4.60379932998954, "tokens_seen": 130220032 }, { "epoch": 0.04, "learning_rate": 0.00048511534603811433, "loss": 3.4426, "theoretical_loss": 4.6034901330182265, "tokens_seen": 130285568 }, { "epoch": 0.04, "learning_rate": 0.0004851053159478435, "loss": 3.685, "theoretical_loss": 4.603181135063394, "tokens_seen": 130351104 }, { "epoch": 0.04, "learning_rate": 0.00048509528585757275, "loss": 3.2137, "theoretical_loss": 4.6028723358969845, "tokens_seen": 130416640 }, { "epoch": 0.04, "learning_rate": 0.0004850852557673019, "loss": 3.3579, "theoretical_loss": 4.602563735291312, "tokens_seen": 130482176 }, { "epoch": 0.04, "learning_rate": 0.0004850752256770311, "loss": 3.3985, "theoretical_loss": 4.602255333019068, "tokens_seen": 130547712 }, { "epoch": 0.04, "learning_rate": 0.00048506519558676024, "loss": 3.4027, "theoretical_loss": 4.6019471288533165, "tokens_seen": 130613248 }, { "epoch": 0.04, "learning_rate": 0.0004850551654964895, "loss": 3.4611, "theoretical_loss": 4.601639122567497, "tokens_seen": 130678784 }, { "epoch": 0.04, "learning_rate": 0.00048504513540621866, "loss": 3.2935, "theoretical_loss": 4.601331313935418, "tokens_seen": 130744320 }, { "epoch": 0.04, "learning_rate": 0.00048503510531594784, "loss": 3.3577, "theoretical_loss": 4.601023702731264, "tokens_seen": 130809856 }, { "epoch": 0.04, "learning_rate": 0.000485025075225677, "loss": 3.3172, "theoretical_loss": 4.600716288729587, "tokens_seen": 130875392 }, { "epoch": 0.04, "learning_rate": 0.00048501504513540625, "loss": 3.2354, "theoretical_loss": 4.600409071705312, "tokens_seen": 130940928 }, { "epoch": 0.04, "learning_rate": 0.0004850050150451354, "loss": 3.406, "theoretical_loss": 4.60010205143373, "tokens_seen": 131006464 }, { "epoch": 0.04, "objective/train/docs_used": 85560, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.442735195159912, "objective/train/theoretical_loss": 4.599795227690505, "objective/train/tokens_used": 151532000, "theoretical_loss": 4.599795227690505, "tokens_seen": 131072000 }, { "epoch": 0.04, "learning_rate": 0.0004849949849548646, "loss": 3.4252, "theoretical_loss": 4.599795227690505, "tokens_seen": 131072000 }, { "epoch": 0.04, "learning_rate": 0.0004849849548645938, "loss": 3.2521, "theoretical_loss": 4.5994886002516635, "tokens_seen": 131137536 }, { "epoch": 0.04, "learning_rate": 0.000484974924774323, "loss": 3.2135, "theoretical_loss": 4.599182168893604, "tokens_seen": 131203072 }, { "epoch": 0.04, "learning_rate": 0.0004849648946840522, "loss": 3.29, "theoretical_loss": 4.598875933393089, "tokens_seen": 131268608 }, { "epoch": 0.04, "learning_rate": 0.00048495486459378134, "loss": 3.4736, "theoretical_loss": 4.5985698935272445, "tokens_seen": 131334144 }, { "epoch": 0.04, "learning_rate": 0.0004849448345035106, "loss": 3.361, "theoretical_loss": 4.598264049073565, "tokens_seen": 131399680 }, { "epoch": 0.04, "learning_rate": 0.0004849348044132397, "loss": 3.1653, "theoretical_loss": 4.597958399809908, "tokens_seen": 131465216 }, { "epoch": 0.04, "learning_rate": 0.00048492477432296894, "loss": 3.0195, "theoretical_loss": 4.59765294551449, "tokens_seen": 131530752 }, { "epoch": 0.04, "learning_rate": 0.0004849147442326981, "loss": 3.4434, "theoretical_loss": 4.597347685965897, "tokens_seen": 131596288 }, { "epoch": 0.04, "learning_rate": 0.0004849047141424273, "loss": 3.579, "theoretical_loss": 4.597042620943069, "tokens_seen": 131661824 }, { "epoch": 0.04, "learning_rate": 0.0004848946840521565, "loss": 3.1773, "theoretical_loss": 4.596737750225311, "tokens_seen": 131727360 }, { "epoch": 0.04, "learning_rate": 0.0004848846539618857, "loss": 3.3462, "theoretical_loss": 4.596433073592289, "tokens_seen": 131792896 }, { "epoch": 0.04, "learning_rate": 0.00048487462387161484, "loss": 3.169, "theoretical_loss": 4.596128590824026, "tokens_seen": 131858432 }, { "epoch": 0.04, "learning_rate": 0.0004848645937813441, "loss": 3.3918, "theoretical_loss": 4.595824301700904, "tokens_seen": 131923968 }, { "epoch": 0.04, "learning_rate": 0.0004848545636910732, "loss": 3.1064, "theoretical_loss": 4.595520206003663, "tokens_seen": 131989504 }, { "epoch": 0.04, "learning_rate": 0.00048484453360080244, "loss": 3.4125, "theoretical_loss": 4.595216303513399, "tokens_seen": 132055040 }, { "epoch": 0.04, "learning_rate": 0.0004848345035105316, "loss": 3.5448, "theoretical_loss": 4.594912594011566, "tokens_seen": 132120576 }, { "epoch": 0.04, "learning_rate": 0.0004848244734202608, "loss": 3.1756, "theoretical_loss": 4.594609077279973, "tokens_seen": 132186112 }, { "epoch": 0.04, "learning_rate": 0.00048481444332999, "loss": 3.3843, "theoretical_loss": 4.594305753100782, "tokens_seen": 132251648 }, { "epoch": 0.04, "learning_rate": 0.00048480441323971917, "loss": 3.4335, "theoretical_loss": 4.594002621256511, "tokens_seen": 132317184 }, { "epoch": 0.04, "learning_rate": 0.00048479438314944835, "loss": 3.4907, "theoretical_loss": 4.59369968153003, "tokens_seen": 132382720 }, { "epoch": 0.04, "learning_rate": 0.0004847843530591776, "loss": 3.3829, "theoretical_loss": 4.593396933704562, "tokens_seen": 132448256 }, { "epoch": 0.04, "learning_rate": 0.0004847743229689067, "loss": 3.4217, "theoretical_loss": 4.593094377563681, "tokens_seen": 132513792 }, { "epoch": 0.04, "learning_rate": 0.00048476429287863594, "loss": 3.5836, "theoretical_loss": 4.592792012891314, "tokens_seen": 132579328 }, { "epoch": 0.04, "learning_rate": 0.00048475426278836507, "loss": 3.1773, "theoretical_loss": 4.592489839471735, "tokens_seen": 132644864 }, { "epoch": 0.04, "objective/train/docs_used": 86181, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.529838800430298, "objective/train/theoretical_loss": 4.592187857089571, "objective/train/tokens_used": 153170400, "theoretical_loss": 4.592187857089571, "tokens_seen": 132710400 }, { "epoch": 0.04, "learning_rate": 0.0004847442326980943, "loss": 3.3219, "theoretical_loss": 4.592187857089571, "tokens_seen": 132710400 }, { "epoch": 0.04, "learning_rate": 0.0004847342026078235, "loss": 3.0821, "theoretical_loss": 4.591886065529795, "tokens_seen": 132775936 }, { "epoch": 0.04, "learning_rate": 0.00048472417251755267, "loss": 3.2018, "theoretical_loss": 4.591584464577728, "tokens_seen": 132841472 }, { "epoch": 0.04, "learning_rate": 0.00048471414242728185, "loss": 3.3234, "theoretical_loss": 4.591283054019041, "tokens_seen": 132907008 }, { "epoch": 0.04, "learning_rate": 0.0004847041123370111, "loss": 3.3791, "theoretical_loss": 4.5909818336397485, "tokens_seen": 132972544 }, { "epoch": 0.04, "learning_rate": 0.0004846940822467402, "loss": 3.2039, "theoretical_loss": 4.590680803226213, "tokens_seen": 133038080 }, { "epoch": 0.04, "learning_rate": 0.00048468405215646945, "loss": 3.4776, "theoretical_loss": 4.590379962565141, "tokens_seen": 133103616 }, { "epoch": 0.04, "learning_rate": 0.0004846740220661986, "loss": 3.337, "theoretical_loss": 4.590079311443583, "tokens_seen": 133169152 }, { "epoch": 0.04, "learning_rate": 0.0004846639919759278, "loss": 3.431, "theoretical_loss": 4.589778849648934, "tokens_seen": 133234688 }, { "epoch": 0.04, "learning_rate": 0.000484653961885657, "loss": 3.3345, "theoretical_loss": 4.589478576968932, "tokens_seen": 133300224 }, { "epoch": 0.04, "learning_rate": 0.00048464393179538617, "loss": 3.4218, "theoretical_loss": 4.589178493191655, "tokens_seen": 133365760 }, { "epoch": 0.04, "learning_rate": 0.00048463390170511535, "loss": 3.4226, "theoretical_loss": 4.588878598105527, "tokens_seen": 133431296 }, { "epoch": 0.04, "learning_rate": 0.00048462387161484453, "loss": 3.3897, "theoretical_loss": 4.588578891499308, "tokens_seen": 133496832 }, { "epoch": 0.04, "learning_rate": 0.0004846138415245737, "loss": 3.3441, "theoretical_loss": 4.588279373162101, "tokens_seen": 133562368 }, { "epoch": 0.04, "learning_rate": 0.00048460381143430295, "loss": 3.1719, "theoretical_loss": 4.587980042883347, "tokens_seen": 133627904 }, { "epoch": 0.04, "learning_rate": 0.0004845937813440321, "loss": 3.5295, "theoretical_loss": 4.587680900452824, "tokens_seen": 133693440 }, { "epoch": 0.04, "learning_rate": 0.0004845837512537613, "loss": 3.3176, "theoretical_loss": 4.587381945660653, "tokens_seen": 133758976 }, { "epoch": 0.04, "learning_rate": 0.00048457372116349044, "loss": 3.1641, "theoretical_loss": 4.587083178297288, "tokens_seen": 133824512 }, { "epoch": 0.04, "learning_rate": 0.0004845636910732197, "loss": 3.1573, "theoretical_loss": 4.5867845981535185, "tokens_seen": 133890048 }, { "epoch": 0.04, "learning_rate": 0.00048455366098294886, "loss": 3.3356, "theoretical_loss": 4.586486205020474, "tokens_seen": 133955584 }, { "epoch": 0.04, "learning_rate": 0.00048454363089267804, "loss": 3.331, "theoretical_loss": 4.586187998689616, "tokens_seen": 134021120 }, { "epoch": 0.04, "learning_rate": 0.0004845336008024072, "loss": 3.464, "theoretical_loss": 4.585889978952741, "tokens_seen": 134086656 }, { "epoch": 0.04, "learning_rate": 0.00048452357071213645, "loss": 3.1701, "theoretical_loss": 4.58559214560198, "tokens_seen": 134152192 }, { "epoch": 0.04, "learning_rate": 0.0004845135406218656, "loss": 3.6266, "theoretical_loss": 4.585294498429796, "tokens_seen": 134217728 }, { "epoch": 0.04, "learning_rate": 0.0004845035105315948, "loss": 3.3988, "theoretical_loss": 4.584997037228986, "tokens_seen": 134283264 }, { "epoch": 0.04, "objective/train/docs_used": 87535, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1884214878082275, "objective/train/theoretical_loss": 4.584699761792674, "objective/train/tokens_used": 154808800, "theoretical_loss": 4.584699761792674, "tokens_seen": 134348800 }, { "epoch": 0.04, "learning_rate": 0.00048449348044132394, "loss": 3.2752, "theoretical_loss": 4.584699761792674, "tokens_seen": 134348800 }, { "epoch": 0.04, "learning_rate": 0.0004844834503510532, "loss": 3.3727, "theoretical_loss": 4.5844026719143205, "tokens_seen": 134414336 }, { "epoch": 0.04, "learning_rate": 0.00048447342026078236, "loss": 3.4484, "theoretical_loss": 4.5841057673877135, "tokens_seen": 134479872 }, { "epoch": 0.04, "learning_rate": 0.00048446339017051154, "loss": 3.429, "theoretical_loss": 4.5838090480069695, "tokens_seen": 134545408 }, { "epoch": 0.04, "learning_rate": 0.0004844533600802407, "loss": 3.1828, "theoretical_loss": 4.5835125135665375, "tokens_seen": 134610944 }, { "epoch": 0.04, "learning_rate": 0.0004844433299899699, "loss": 3.2808, "theoretical_loss": 4.583216163861191, "tokens_seen": 134676480 }, { "epoch": 0.04, "learning_rate": 0.0004844332998996991, "loss": 3.4666, "theoretical_loss": 4.58291999868603, "tokens_seen": 134742016 }, { "epoch": 0.04, "learning_rate": 0.0004844232698094283, "loss": 3.0757, "theoretical_loss": 4.582624017836489, "tokens_seen": 134807552 }, { "epoch": 0.04, "learning_rate": 0.00048441323971915745, "loss": 3.5777, "theoretical_loss": 4.582328221108318, "tokens_seen": 134873088 }, { "epoch": 0.04, "learning_rate": 0.0004844032096288867, "loss": 3.1727, "theoretical_loss": 4.5820326082976, "tokens_seen": 134938624 }, { "epoch": 0.04, "learning_rate": 0.00048439317953861586, "loss": 3.6033, "theoretical_loss": 4.581737179200739, "tokens_seen": 135004160 }, { "epoch": 0.04, "learning_rate": 0.00048438314944834504, "loss": 3.3186, "theoretical_loss": 4.581441933614466, "tokens_seen": 135069696 }, { "epoch": 0.04, "learning_rate": 0.0004843731193580742, "loss": 3.3087, "theoretical_loss": 4.581146871335832, "tokens_seen": 135135232 }, { "epoch": 0.04, "learning_rate": 0.0004843630892678034, "loss": 3.0668, "theoretical_loss": 4.580851992162214, "tokens_seen": 135200768 }, { "epoch": 0.04, "learning_rate": 0.0004843530591775326, "loss": 3.2295, "theoretical_loss": 4.5805572958913086, "tokens_seen": 135266304 }, { "epoch": 0.04, "learning_rate": 0.0004843430290872618, "loss": 3.5356, "theoretical_loss": 4.580262782321135, "tokens_seen": 135331840 }, { "epoch": 0.04, "learning_rate": 0.00048433299899699095, "loss": 3.3037, "theoretical_loss": 4.579968451250032, "tokens_seen": 135397376 }, { "epoch": 0.04, "learning_rate": 0.0004843229689067202, "loss": 3.2675, "theoretical_loss": 4.579674302476661, "tokens_seen": 135462912 }, { "epoch": 0.04, "learning_rate": 0.0004843129388164493, "loss": 3.2687, "theoretical_loss": 4.579380335800001, "tokens_seen": 135528448 }, { "epoch": 0.04, "learning_rate": 0.00048430290872617855, "loss": 3.5504, "theoretical_loss": 4.579086551019348, "tokens_seen": 135593984 }, { "epoch": 0.04, "learning_rate": 0.00048429287863590773, "loss": 3.4594, "theoretical_loss": 4.5787929479343195, "tokens_seen": 135659520 }, { "epoch": 0.04, "learning_rate": 0.0004842828485456369, "loss": 3.2982, "theoretical_loss": 4.578499526344848, "tokens_seen": 135725056 }, { "epoch": 0.04, "learning_rate": 0.0004842728184553661, "loss": 3.1927, "theoretical_loss": 4.578206286051184, "tokens_seen": 135790592 }, { "epoch": 0.04, "learning_rate": 0.00048426278836509527, "loss": 3.3044, "theoretical_loss": 4.5779132268538945, "tokens_seen": 135856128 }, { "epoch": 0.04, "learning_rate": 0.00048425275827482445, "loss": 3.4211, "theoretical_loss": 4.577620348553859, "tokens_seen": 135921664 }, { "epoch": 0.04, "objective/train/docs_used": 88189, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8670549392700195, "objective/train/theoretical_loss": 4.577327650952276, "objective/train/tokens_used": 156447200, "theoretical_loss": 4.577327650952276, "tokens_seen": 135987200 }, { "epoch": 0.04, "learning_rate": 0.0004842427281845537, "loss": 3.3964, "theoretical_loss": 4.577327650952276, "tokens_seen": 135987200 }, { "epoch": 0.04, "learning_rate": 0.00048423269809428287, "loss": 3.0862, "theoretical_loss": 4.5770351338506545, "tokens_seen": 136052736 }, { "epoch": 0.04, "learning_rate": 0.00048422266800401205, "loss": 3.0829, "theoretical_loss": 4.57674279705082, "tokens_seen": 136118272 }, { "epoch": 0.04, "learning_rate": 0.0004842126379137413, "loss": 3.4358, "theoretical_loss": 4.57645064035491, "tokens_seen": 136183808 }, { "epoch": 0.04, "learning_rate": 0.0004842026078234704, "loss": 3.4142, "theoretical_loss": 4.576158663565371, "tokens_seen": 136249344 }, { "epoch": 0.04, "learning_rate": 0.00048419257773319965, "loss": 3.3851, "theoretical_loss": 4.575866866484967, "tokens_seen": 136314880 }, { "epoch": 0.04, "learning_rate": 0.0004841825476429288, "loss": 3.4905, "theoretical_loss": 4.575575248916767, "tokens_seen": 136380416 }, { "epoch": 0.04, "learning_rate": 0.000484172517552658, "loss": 3.2962, "theoretical_loss": 4.575283810664155, "tokens_seen": 136445952 }, { "epoch": 0.04, "learning_rate": 0.0004841624874623872, "loss": 3.3953, "theoretical_loss": 4.574992551530822, "tokens_seen": 136511488 }, { "epoch": 0.04, "learning_rate": 0.00048415245737211637, "loss": 3.4929, "theoretical_loss": 4.574701471320768, "tokens_seen": 136577024 }, { "epoch": 0.04, "learning_rate": 0.00048414242728184555, "loss": 3.2127, "theoretical_loss": 4.574410569838304, "tokens_seen": 136642560 }, { "epoch": 0.04, "learning_rate": 0.00048413239719157473, "loss": 3.3957, "theoretical_loss": 4.574119846888045, "tokens_seen": 136708096 }, { "epoch": 0.04, "learning_rate": 0.0004841223671013039, "loss": 3.4099, "theoretical_loss": 4.573829302274915, "tokens_seen": 136773632 }, { "epoch": 0.04, "learning_rate": 0.00048411233701103315, "loss": 3.4686, "theoretical_loss": 4.573538935804146, "tokens_seen": 136839168 }, { "epoch": 0.04, "learning_rate": 0.0004841023069207623, "loss": 3.0463, "theoretical_loss": 4.573248747281273, "tokens_seen": 136904704 }, { "epoch": 0.04, "learning_rate": 0.0004840922768304915, "loss": 3.7113, "theoretical_loss": 4.5729587365121365, "tokens_seen": 136970240 }, { "epoch": 0.04, "learning_rate": 0.00048408224674022064, "loss": 3.3974, "theoretical_loss": 4.572668903302886, "tokens_seen": 137035776 }, { "epoch": 0.04, "learning_rate": 0.0004840722166499499, "loss": 3.184, "theoretical_loss": 4.572379247459969, "tokens_seen": 137101312 }, { "epoch": 0.04, "learning_rate": 0.00048406218655967906, "loss": 3.4432, "theoretical_loss": 4.57208976879014, "tokens_seen": 137166848 }, { "epoch": 0.04, "learning_rate": 0.00048405215646940824, "loss": 3.5294, "theoretical_loss": 4.571800467100456, "tokens_seen": 137232384 }, { "epoch": 0.04, "learning_rate": 0.0004840421263791374, "loss": 3.0596, "theoretical_loss": 4.5715113421982725, "tokens_seen": 137297920 }, { "epoch": 0.04, "learning_rate": 0.00048403209628886665, "loss": 3.3131, "theoretical_loss": 4.571222393891253, "tokens_seen": 137363456 }, { "epoch": 0.04, "learning_rate": 0.0004840220661985958, "loss": 3.51, "theoretical_loss": 4.570933621987356, "tokens_seen": 137428992 }, { "epoch": 0.04, "learning_rate": 0.000484012036108325, "loss": 3.4489, "theoretical_loss": 4.570645026294844, "tokens_seen": 137494528 }, { "epoch": 0.04, "learning_rate": 0.00048400200601805414, "loss": 3.3655, "theoretical_loss": 4.570356606622278, "tokens_seen": 137560064 }, { "epoch": 0.04, "objective/train/docs_used": 88907, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2466330528259277, "objective/train/theoretical_loss": 4.570068362778516, "objective/train/tokens_used": 158085600, "theoretical_loss": 4.570068362778516, "tokens_seen": 137625600 }, { "epoch": 0.04, "learning_rate": 0.0004839919759277834, "loss": 3.4146, "theoretical_loss": 4.570068362778516, "tokens_seen": 137625600 }, { "epoch": 0.04, "learning_rate": 0.00048398194583751256, "loss": 3.3105, "theoretical_loss": 4.569780294572718, "tokens_seen": 137691136 }, { "epoch": 0.04, "learning_rate": 0.00048397191574724174, "loss": 3.5781, "theoretical_loss": 4.569492401814339, "tokens_seen": 137756672 }, { "epoch": 0.04, "learning_rate": 0.0004839618856569709, "loss": 3.047, "theoretical_loss": 4.569204684313133, "tokens_seen": 137822208 }, { "epoch": 0.04, "learning_rate": 0.0004839518555667001, "loss": 3.4353, "theoretical_loss": 4.568917141879149, "tokens_seen": 137887744 }, { "epoch": 0.04, "learning_rate": 0.0004839418254764293, "loss": 3.1746, "theoretical_loss": 4.568629774322736, "tokens_seen": 137953280 }, { "epoch": 0.04, "learning_rate": 0.0004839317953861585, "loss": 3.4604, "theoretical_loss": 4.568342581454532, "tokens_seen": 138018816 }, { "epoch": 0.04, "learning_rate": 0.00048392176529588765, "loss": 3.5407, "theoretical_loss": 4.568055563085476, "tokens_seen": 138084352 }, { "epoch": 0.04, "learning_rate": 0.0004839117352056169, "loss": 3.2899, "theoretical_loss": 4.567768719026797, "tokens_seen": 138149888 }, { "epoch": 0.04, "learning_rate": 0.00048390170511534606, "loss": 3.3983, "theoretical_loss": 4.567482049090019, "tokens_seen": 138215424 }, { "epoch": 0.04, "learning_rate": 0.00048389167502507524, "loss": 3.2304, "theoretical_loss": 4.567195553086961, "tokens_seen": 138280960 }, { "epoch": 0.04, "learning_rate": 0.0004838816449348044, "loss": 3.3718, "theoretical_loss": 4.566909230829729, "tokens_seen": 138346496 }, { "epoch": 0.04, "learning_rate": 0.0004838716148445336, "loss": 3.4189, "theoretical_loss": 4.566623082130729, "tokens_seen": 138412032 }, { "epoch": 0.04, "learning_rate": 0.0004838615847542628, "loss": 3.197, "theoretical_loss": 4.566337106802651, "tokens_seen": 138477568 }, { "epoch": 0.04, "learning_rate": 0.000483851554663992, "loss": 3.4958, "theoretical_loss": 4.56605130465848, "tokens_seen": 138543104 }, { "epoch": 0.04, "learning_rate": 0.00048384152457372115, "loss": 3.245, "theoretical_loss": 4.565765675511487, "tokens_seen": 138608640 }, { "epoch": 0.04, "learning_rate": 0.0004838314944834504, "loss": 3.3454, "theoretical_loss": 4.565480219175237, "tokens_seen": 138674176 }, { "epoch": 0.04, "learning_rate": 0.0004838214643931795, "loss": 3.5074, "theoretical_loss": 4.56519493546358, "tokens_seen": 138739712 }, { "epoch": 0.04, "learning_rate": 0.00048381143430290875, "loss": 3.5971, "theoretical_loss": 4.56490982419066, "tokens_seen": 138805248 }, { "epoch": 0.04, "learning_rate": 0.00048380140421263793, "loss": 3.3556, "theoretical_loss": 4.564624885170902, "tokens_seen": 138870784 }, { "epoch": 0.04, "learning_rate": 0.0004837913741223671, "loss": 3.4486, "theoretical_loss": 4.564340118219022, "tokens_seen": 138936320 }, { "epoch": 0.04, "learning_rate": 0.0004837813440320963, "loss": 3.3375, "theoretical_loss": 4.56405552315002, "tokens_seen": 139001856 }, { "epoch": 0.04, "learning_rate": 0.00048377131394182547, "loss": 3.5742, "theoretical_loss": 4.563771099779187, "tokens_seen": 139067392 }, { "epoch": 0.04, "learning_rate": 0.00048376128385155465, "loss": 3.402, "theoretical_loss": 4.563486847922093, "tokens_seen": 139132928 }, { "epoch": 0.04, "learning_rate": 0.0004837512537612839, "loss": 3.221, "theoretical_loss": 4.563202767394597, "tokens_seen": 139198464 }, { "epoch": 0.04, "objective/train/docs_used": 90119, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3730666637420654, "objective/train/theoretical_loss": 4.562918858012843, "objective/train/tokens_used": 159724000, "theoretical_loss": 4.562918858012843, "tokens_seen": 139264000 }, { "epoch": 0.04, "learning_rate": 0.000483741223671013, "loss": 3.459, "theoretical_loss": 4.562918858012843, "tokens_seen": 139264000 }, { "epoch": 0.04, "learning_rate": 0.00048373119358074225, "loss": 3.4475, "theoretical_loss": 4.562635119593255, "tokens_seen": 139329536 }, { "epoch": 0.04, "learning_rate": 0.00048372116349047143, "loss": 3.523, "theoretical_loss": 4.562351551952542, "tokens_seen": 139395072 }, { "epoch": 0.04, "learning_rate": 0.0004837111334002006, "loss": 3.434, "theoretical_loss": 4.5620681549076965, "tokens_seen": 139460608 }, { "epoch": 0.04, "learning_rate": 0.0004837011033099298, "loss": 3.4291, "theoretical_loss": 4.561784928275992, "tokens_seen": 139526144 }, { "epoch": 0.04, "learning_rate": 0.000483691073219659, "loss": 3.3811, "theoretical_loss": 4.561501871874984, "tokens_seen": 139591680 }, { "epoch": 0.04, "learning_rate": 0.00048368104312938816, "loss": 3.3562, "theoretical_loss": 4.561218985522507, "tokens_seen": 139657216 }, { "epoch": 0.04, "learning_rate": 0.0004836710130391174, "loss": 3.7033, "theoretical_loss": 4.560936269036679, "tokens_seen": 139722752 }, { "epoch": 0.04, "learning_rate": 0.0004836609829488465, "loss": 3.4809, "theoretical_loss": 4.560653722235895, "tokens_seen": 139788288 }, { "epoch": 0.04, "learning_rate": 0.00048365095285857575, "loss": 3.6892, "theoretical_loss": 4.560371344938831, "tokens_seen": 139853824 }, { "epoch": 0.04, "learning_rate": 0.0004836409227683049, "loss": 3.3982, "theoretical_loss": 4.560089136964439, "tokens_seen": 139919360 }, { "epoch": 0.04, "learning_rate": 0.0004836308926780341, "loss": 3.7439, "theoretical_loss": 4.559807098131953, "tokens_seen": 139984896 }, { "epoch": 0.04, "learning_rate": 0.0004836208625877633, "loss": 3.2185, "theoretical_loss": 4.559525228260882, "tokens_seen": 140050432 }, { "epoch": 0.04, "learning_rate": 0.0004836108324974925, "loss": 3.2183, "theoretical_loss": 4.559243527171011, "tokens_seen": 140115968 }, { "epoch": 0.04, "learning_rate": 0.00048360080240722166, "loss": 3.4715, "theoretical_loss": 4.558961994682403, "tokens_seen": 140181504 }, { "epoch": 0.04, "learning_rate": 0.00048359077231695084, "loss": 3.5439, "theoretical_loss": 4.558680630615397, "tokens_seen": 140247040 }, { "epoch": 0.04, "learning_rate": 0.00048358074222668, "loss": 3.4811, "theoretical_loss": 4.558399434790607, "tokens_seen": 140312576 }, { "epoch": 0.04, "learning_rate": 0.00048357071213640926, "loss": 3.4954, "theoretical_loss": 4.558118407028921, "tokens_seen": 140378112 }, { "epoch": 0.04, "learning_rate": 0.0004835606820461384, "loss": 3.3851, "theoretical_loss": 4.557837547151502, "tokens_seen": 140443648 }, { "epoch": 0.04, "learning_rate": 0.0004835506519558676, "loss": 3.3218, "theoretical_loss": 4.557556854979786, "tokens_seen": 140509184 }, { "epoch": 0.04, "learning_rate": 0.0004835406218655968, "loss": 3.4642, "theoretical_loss": 4.5572763303354815, "tokens_seen": 140574720 }, { "epoch": 0.04, "learning_rate": 0.000483530591775326, "loss": 3.237, "theoretical_loss": 4.556995973040574, "tokens_seen": 140640256 }, { "epoch": 0.04, "learning_rate": 0.00048352056168505516, "loss": 3.4711, "theoretical_loss": 4.556715782917314, "tokens_seen": 140705792 }, { "epoch": 0.04, "learning_rate": 0.00048351053159478434, "loss": 3.3196, "theoretical_loss": 4.556435759788229, "tokens_seen": 140771328 }, { "epoch": 0.04, "learning_rate": 0.0004835005015045135, "loss": 3.2629, "theoretical_loss": 4.556155903476114, "tokens_seen": 140836864 }, { "epoch": 0.04, "objective/train/docs_used": 90792, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.9938900470733643, "objective/train/theoretical_loss": 4.555876213804037, "objective/train/tokens_used": 161362400, "theoretical_loss": 4.555876213804037, "tokens_seen": 140902400 }, { "epoch": 0.04, "learning_rate": 0.00048349047141424276, "loss": 3.6836, "theoretical_loss": 4.555876213804037, "tokens_seen": 140902400 }, { "epoch": 0.04, "learning_rate": 0.00048348044132397194, "loss": 3.4205, "theoretical_loss": 4.555596690595333, "tokens_seen": 140967936 }, { "epoch": 0.04, "learning_rate": 0.0004834704112337011, "loss": 3.3216, "theoretical_loss": 4.555317333673611, "tokens_seen": 141033472 }, { "epoch": 0.04, "learning_rate": 0.0004834603811434303, "loss": 3.247, "theoretical_loss": 4.555038142862742, "tokens_seen": 141099008 }, { "epoch": 0.04, "learning_rate": 0.0004834503510531595, "loss": 3.3618, "theoretical_loss": 4.5547591179868725, "tokens_seen": 141164544 }, { "epoch": 0.04, "learning_rate": 0.0004834403209628887, "loss": 3.3634, "theoretical_loss": 4.554480258870409, "tokens_seen": 141230080 }, { "epoch": 0.04, "learning_rate": 0.00048343029087261785, "loss": 3.4449, "theoretical_loss": 4.554201565338033, "tokens_seen": 141295616 }, { "epoch": 0.04, "learning_rate": 0.0004834202607823471, "loss": 3.212, "theoretical_loss": 4.5539230372146875, "tokens_seen": 141361152 }, { "epoch": 0.04, "learning_rate": 0.00048341023069207626, "loss": 3.2166, "theoretical_loss": 4.553644674325584, "tokens_seen": 141426688 }, { "epoch": 0.04, "learning_rate": 0.00048340020060180544, "loss": 3.4181, "theoretical_loss": 4.553366476496198, "tokens_seen": 141492224 }, { "epoch": 0.04, "learning_rate": 0.0004833901705115346, "loss": 3.2078, "theoretical_loss": 4.553088443552269, "tokens_seen": 141557760 }, { "epoch": 0.04, "learning_rate": 0.0004833801404212638, "loss": 3.3662, "theoretical_loss": 4.552810575319806, "tokens_seen": 141623296 }, { "epoch": 0.04, "learning_rate": 0.000483370110330993, "loss": 3.2753, "theoretical_loss": 4.552532871625077, "tokens_seen": 141688832 }, { "epoch": 0.04, "learning_rate": 0.0004833600802407222, "loss": 3.4694, "theoretical_loss": 4.5522553322946155, "tokens_seen": 141754368 }, { "epoch": 0.04, "learning_rate": 0.00048335005015045135, "loss": 3.3584, "theoretical_loss": 4.551977957155217, "tokens_seen": 141819904 }, { "epoch": 0.04, "learning_rate": 0.0004833400200601806, "loss": 3.3498, "theoretical_loss": 4.5517007460339425, "tokens_seen": 141885440 }, { "epoch": 0.04, "learning_rate": 0.0004833299899699097, "loss": 3.4718, "theoretical_loss": 4.551423698758111, "tokens_seen": 141950976 }, { "epoch": 0.04, "learning_rate": 0.00048331995987963895, "loss": 3.4992, "theoretical_loss": 4.551146815155304, "tokens_seen": 142016512 }, { "epoch": 0.04, "learning_rate": 0.00048330992978936813, "loss": 3.1915, "theoretical_loss": 4.550870095053366, "tokens_seen": 142082048 }, { "epoch": 0.04, "learning_rate": 0.0004832998996990973, "loss": 3.4141, "theoretical_loss": 4.550593538280398, "tokens_seen": 142147584 }, { "epoch": 0.04, "learning_rate": 0.0004832898696088265, "loss": 3.4559, "theoretical_loss": 4.550317144664766, "tokens_seen": 142213120 }, { "epoch": 0.04, "learning_rate": 0.00048327983951855567, "loss": 3.1613, "theoretical_loss": 4.55004091403509, "tokens_seen": 142278656 }, { "epoch": 0.04, "learning_rate": 0.00048326980942828485, "loss": 3.4072, "theoretical_loss": 4.5497648462202545, "tokens_seen": 142344192 }, { "epoch": 0.04, "learning_rate": 0.0004832597793380141, "loss": 3.4498, "theoretical_loss": 4.549488941049397, "tokens_seen": 142409728 }, { "epoch": 0.04, "learning_rate": 0.0004832497492477432, "loss": 3.1324, "theoretical_loss": 4.549213198351914, "tokens_seen": 142475264 }, { "epoch": 0.04, "objective/train/docs_used": 92175, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.230867624282837, "objective/train/theoretical_loss": 4.548937617957463, "objective/train/tokens_used": 163000800, "theoretical_loss": 4.548937617957463, "tokens_seen": 142540800 }, { "epoch": 0.04, "learning_rate": 0.00048323971915747245, "loss": 3.3699, "theoretical_loss": 4.548937617957463, "tokens_seen": 142540800 }, { "epoch": 0.04, "learning_rate": 0.00048322968906720163, "loss": 3.3031, "theoretical_loss": 4.548662199695954, "tokens_seen": 142606336 }, { "epoch": 0.04, "learning_rate": 0.0004832196589769308, "loss": 3.6595, "theoretical_loss": 4.548386943397556, "tokens_seen": 142671872 }, { "epoch": 0.04, "learning_rate": 0.00048320962888666, "loss": 3.518, "theoretical_loss": 4.548111848892693, "tokens_seen": 142737408 }, { "epoch": 0.04, "learning_rate": 0.0004831995987963892, "loss": 3.1302, "theoretical_loss": 4.547836916012042, "tokens_seen": 142802944 }, { "epoch": 0.04, "learning_rate": 0.00048318956870611836, "loss": 3.4835, "theoretical_loss": 4.547562144586539, "tokens_seen": 142868480 }, { "epoch": 0.04, "learning_rate": 0.0004831795386158476, "loss": 3.4292, "theoretical_loss": 4.547287534447372, "tokens_seen": 142934016 }, { "epoch": 0.04, "learning_rate": 0.0004831695085255767, "loss": 3.3998, "theoretical_loss": 4.5470130854259825, "tokens_seen": 142999552 }, { "epoch": 0.04, "learning_rate": 0.00048315947843530595, "loss": 3.5444, "theoretical_loss": 4.546738797354065, "tokens_seen": 143065088 }, { "epoch": 0.04, "learning_rate": 0.0004831494483450351, "loss": 3.3672, "theoretical_loss": 4.546464670063569, "tokens_seen": 143130624 }, { "epoch": 0.04, "learning_rate": 0.0004831394182547643, "loss": 3.3033, "theoretical_loss": 4.546190703386695, "tokens_seen": 143196160 }, { "epoch": 0.04, "learning_rate": 0.0004831293881644935, "loss": 3.2802, "theoretical_loss": 4.545916897155894, "tokens_seen": 143261696 }, { "epoch": 0.04, "learning_rate": 0.0004831193580742227, "loss": 3.2078, "theoretical_loss": 4.54564325120387, "tokens_seen": 143327232 }, { "epoch": 0.04, "learning_rate": 0.00048310932798395186, "loss": 3.369, "theoretical_loss": 4.545369765363578, "tokens_seen": 143392768 }, { "epoch": 0.04, "learning_rate": 0.00048309929789368104, "loss": 3.0522, "theoretical_loss": 4.545096439468223, "tokens_seen": 143458304 }, { "epoch": 0.04, "learning_rate": 0.0004830892678034102, "loss": 3.457, "theoretical_loss": 4.544823273351257, "tokens_seen": 143523840 }, { "epoch": 0.04, "learning_rate": 0.00048307923771313946, "loss": 3.2289, "theoretical_loss": 4.544550266846388, "tokens_seen": 143589376 }, { "epoch": 0.04, "learning_rate": 0.0004830692076228686, "loss": 3.491, "theoretical_loss": 4.544277419787566, "tokens_seen": 143654912 }, { "epoch": 0.04, "learning_rate": 0.0004830591775325978, "loss": 3.273, "theoretical_loss": 4.544004732008993, "tokens_seen": 143720448 }, { "epoch": 0.04, "learning_rate": 0.000483049147442327, "loss": 3.4575, "theoretical_loss": 4.543732203345119, "tokens_seen": 143785984 }, { "epoch": 0.04, "learning_rate": 0.0004830391173520562, "loss": 3.4134, "theoretical_loss": 4.543459833630639, "tokens_seen": 143851520 }, { "epoch": 0.04, "learning_rate": 0.00048302908726178536, "loss": 3.1865, "theoretical_loss": 4.543187622700497, "tokens_seen": 143917056 }, { "epoch": 0.04, "learning_rate": 0.00048301905717151454, "loss": 3.4901, "theoretical_loss": 4.542915570389884, "tokens_seen": 143982592 }, { "epoch": 0.04, "learning_rate": 0.0004830090270812437, "loss": 3.3153, "theoretical_loss": 4.542643676534234, "tokens_seen": 144048128 }, { "epoch": 0.04, "learning_rate": 0.00048299899699097296, "loss": 3.4697, "theoretical_loss": 4.542371940969231, "tokens_seen": 144113664 }, { "epoch": 0.04, "objective/train/docs_used": 92798, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.180187702178955, "objective/train/theoretical_loss": 4.542100363530799, "objective/train/tokens_used": 164639200, "theoretical_loss": 4.542100363530799, "tokens_seen": 144179200 }, { "epoch": 0.04, "learning_rate": 0.0004829889669007021, "loss": 3.4208, "theoretical_loss": 4.542100363530799, "tokens_seen": 144179200 }, { "epoch": 0.04, "learning_rate": 0.0004829789368104313, "loss": 3.4833, "theoretical_loss": 4.54182894405511, "tokens_seen": 144244736 }, { "epoch": 0.04, "learning_rate": 0.00048296890672016045, "loss": 3.5622, "theoretical_loss": 4.5415576823785795, "tokens_seen": 144310272 }, { "epoch": 0.04, "learning_rate": 0.0004829588766298897, "loss": 3.3664, "theoretical_loss": 4.541286578337866, "tokens_seen": 144375808 }, { "epoch": 0.04, "learning_rate": 0.00048294884653961887, "loss": 3.3885, "theoretical_loss": 4.541015631769872, "tokens_seen": 144441344 }, { "epoch": 0.04, "learning_rate": 0.00048293881644934805, "loss": 3.4071, "theoretical_loss": 4.5407448425117405, "tokens_seen": 144506880 }, { "epoch": 0.04, "learning_rate": 0.00048292878635907723, "loss": 3.2401, "theoretical_loss": 4.540474210400859, "tokens_seen": 144572416 }, { "epoch": 0.04, "learning_rate": 0.00048291875626880646, "loss": 3.3437, "theoretical_loss": 4.540203735274855, "tokens_seen": 144637952 }, { "epoch": 0.04, "learning_rate": 0.0004829087261785356, "loss": 3.2116, "theoretical_loss": 4.5399334169716, "tokens_seen": 144703488 }, { "epoch": 0.04, "learning_rate": 0.0004828986960882648, "loss": 3.6115, "theoretical_loss": 4.539663255329202, "tokens_seen": 144769024 }, { "epoch": 0.04, "learning_rate": 0.00048288866599799395, "loss": 3.0955, "theoretical_loss": 4.539393250186015, "tokens_seen": 144834560 }, { "epoch": 0.04, "learning_rate": 0.0004828786359077232, "loss": 3.4351, "theoretical_loss": 4.539123401380625, "tokens_seen": 144900096 }, { "epoch": 0.04, "learning_rate": 0.00048286860581745237, "loss": 3.5045, "theoretical_loss": 4.538853708751866, "tokens_seen": 144965632 }, { "epoch": 0.04, "learning_rate": 0.00048285857572718155, "loss": 3.3471, "theoretical_loss": 4.538584172138804, "tokens_seen": 145031168 }, { "epoch": 0.04, "learning_rate": 0.00048284854563691073, "loss": 3.2684, "theoretical_loss": 4.538314791380748, "tokens_seen": 145096704 }, { "epoch": 0.04, "learning_rate": 0.0004828385155466399, "loss": 3.2516, "theoretical_loss": 4.538045566317242, "tokens_seen": 145162240 }, { "epoch": 0.04, "learning_rate": 0.0004828284854563691, "loss": 3.4594, "theoretical_loss": 4.537776496788071, "tokens_seen": 145227776 }, { "epoch": 0.04, "learning_rate": 0.00048281845536609833, "loss": 3.603, "theoretical_loss": 4.537507582633253, "tokens_seen": 145293312 }, { "epoch": 0.04, "learning_rate": 0.00048280842527582746, "loss": 3.2846, "theoretical_loss": 4.537238823693045, "tokens_seen": 145358848 }, { "epoch": 0.04, "learning_rate": 0.0004827983951855567, "loss": 3.4105, "theoretical_loss": 4.536970219807939, "tokens_seen": 145424384 }, { "epoch": 0.04, "learning_rate": 0.0004827883650952858, "loss": 3.3511, "theoretical_loss": 4.536701770818665, "tokens_seen": 145489920 }, { "epoch": 0.04, "learning_rate": 0.00048277833500501505, "loss": 3.2664, "theoretical_loss": 4.536433476566185, "tokens_seen": 145555456 }, { "epoch": 0.04, "learning_rate": 0.00048276830491474423, "loss": 3.2244, "theoretical_loss": 4.536165336891699, "tokens_seen": 145620992 }, { "epoch": 0.04, "learning_rate": 0.0004827582748244734, "loss": 3.5109, "theoretical_loss": 4.535897351636638, "tokens_seen": 145686528 }, { "epoch": 0.04, "learning_rate": 0.0004827482447342026, "loss": 3.509, "theoretical_loss": 4.53562952064267, "tokens_seen": 145752064 }, { "epoch": 0.04, "objective/train/docs_used": 94036, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4322946071624756, "objective/train/theoretical_loss": 4.535361843751696, "objective/train/tokens_used": 166277600, "theoretical_loss": 4.535361843751696, "tokens_seen": 145817600 }, { "epoch": 0.04, "learning_rate": 0.00048273821464393183, "loss": 3.2137, "theoretical_loss": 4.535361843751696, "tokens_seen": 145817600 }, { "epoch": 0.04, "learning_rate": 0.000482728184553661, "loss": 3.0617, "theoretical_loss": 4.535094320805847, "tokens_seen": 145883136 }, { "epoch": 0.04, "learning_rate": 0.0004827181544633902, "loss": 3.4118, "theoretical_loss": 4.534826951647489, "tokens_seen": 145948672 }, { "epoch": 0.04, "learning_rate": 0.0004827081243731194, "loss": 3.3963, "theoretical_loss": 4.5345597361192205, "tokens_seen": 146014208 }, { "epoch": 0.04, "learning_rate": 0.00048269809428284856, "loss": 3.4053, "theoretical_loss": 4.53429267406387, "tokens_seen": 146079744 }, { "epoch": 0.04, "learning_rate": 0.0004826880641925778, "loss": 3.1817, "theoretical_loss": 4.5340257653244995, "tokens_seen": 146145280 }, { "epoch": 0.04, "learning_rate": 0.0004826780341023069, "loss": 3.6942, "theoretical_loss": 4.5337590097444, "tokens_seen": 146210816 }, { "epoch": 0.04, "learning_rate": 0.00048266800401203615, "loss": 3.4861, "theoretical_loss": 4.533492407167093, "tokens_seen": 146276352 }, { "epoch": 0.04, "learning_rate": 0.0004826579739217653, "loss": 3.1617, "theoretical_loss": 4.53322595743633, "tokens_seen": 146341888 }, { "epoch": 0.04, "learning_rate": 0.0004826479438314945, "loss": 3.1373, "theoretical_loss": 4.5329596603960916, "tokens_seen": 146407424 }, { "epoch": 0.04, "learning_rate": 0.0004826379137412237, "loss": 3.1347, "theoretical_loss": 4.53269351589059, "tokens_seen": 146472960 }, { "epoch": 0.04, "learning_rate": 0.0004826278836509529, "loss": 3.3526, "theoretical_loss": 4.532427523764261, "tokens_seen": 146538496 }, { "epoch": 0.04, "learning_rate": 0.00048261785356068206, "loss": 3.2586, "theoretical_loss": 4.532161683861773, "tokens_seen": 146604032 }, { "epoch": 0.04, "learning_rate": 0.00048260782347041124, "loss": 3.3128, "theoretical_loss": 4.5318959960280205, "tokens_seen": 146669568 }, { "epoch": 0.04, "learning_rate": 0.0004825977933801404, "loss": 3.39, "theoretical_loss": 4.531630460108125, "tokens_seen": 146735104 }, { "epoch": 0.04, "learning_rate": 0.00048258776328986966, "loss": 3.4165, "theoretical_loss": 4.531365075947434, "tokens_seen": 146800640 }, { "epoch": 0.04, "learning_rate": 0.0004825777331995988, "loss": 3.2288, "theoretical_loss": 4.531099843391524, "tokens_seen": 146866176 }, { "epoch": 0.04, "learning_rate": 0.000482567703109328, "loss": 3.3832, "theoretical_loss": 4.5308347622861955, "tokens_seen": 146931712 }, { "epoch": 0.04, "learning_rate": 0.0004825576730190572, "loss": 3.6141, "theoretical_loss": 4.5305698324774735, "tokens_seen": 146997248 }, { "epoch": 0.04, "learning_rate": 0.0004825476429287864, "loss": 3.4217, "theoretical_loss": 4.530305053811611, "tokens_seen": 147062784 }, { "epoch": 0.04, "learning_rate": 0.00048253761283851556, "loss": 3.2163, "theoretical_loss": 4.530040426135084, "tokens_seen": 147128320 }, { "epoch": 0.04, "learning_rate": 0.00048252758274824474, "loss": 3.3166, "theoretical_loss": 4.529775949294593, "tokens_seen": 147193856 }, { "epoch": 0.04, "learning_rate": 0.0004825175526579739, "loss": 3.3181, "theoretical_loss": 4.529511623137061, "tokens_seen": 147259392 }, { "epoch": 0.04, "learning_rate": 0.00048250752256770316, "loss": 3.2954, "theoretical_loss": 4.529247447509637, "tokens_seen": 147324928 }, { "epoch": 0.04, "learning_rate": 0.0004824974924774323, "loss": 3.2769, "theoretical_loss": 4.528983422259691, "tokens_seen": 147390464 }, { "epoch": 0.04, "objective/train/docs_used": 94726, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4257872104644775, "objective/train/theoretical_loss": 4.528719547234816, "objective/train/tokens_used": 167916000, "theoretical_loss": 4.528719547234816, "tokens_seen": 147456000 }, { "epoch": 0.04, "learning_rate": 0.0004824874623871615, "loss": 3.3119, "theoretical_loss": 4.528719547234816, "tokens_seen": 147456000 }, { "epoch": 0.04, "learning_rate": 0.00048247743229689065, "loss": 3.3536, "theoretical_loss": 4.528455822282828, "tokens_seen": 147521536 }, { "epoch": 0.04, "learning_rate": 0.0004824674022066199, "loss": 3.4758, "theoretical_loss": 4.528192247251763, "tokens_seen": 147587072 }, { "epoch": 0.04, "learning_rate": 0.00048245737211634907, "loss": 3.1317, "theoretical_loss": 4.52792882198988, "tokens_seen": 147652608 }, { "epoch": 0.04, "learning_rate": 0.00048244734202607825, "loss": 3.3368, "theoretical_loss": 4.527665546345656, "tokens_seen": 147718144 }, { "epoch": 0.04, "learning_rate": 0.00048243731193580743, "loss": 3.6421, "theoretical_loss": 4.5274024201677925, "tokens_seen": 147783680 }, { "epoch": 0.04, "learning_rate": 0.00048242728184553666, "loss": 3.0839, "theoretical_loss": 4.527139443305209, "tokens_seen": 147849216 }, { "epoch": 0.04, "learning_rate": 0.0004824172517552658, "loss": 3.4025, "theoretical_loss": 4.526876615607042, "tokens_seen": 147914752 }, { "epoch": 0.04, "learning_rate": 0.000482407221664995, "loss": 3.1217, "theoretical_loss": 4.526613936922654, "tokens_seen": 147980288 }, { "epoch": 0.04, "learning_rate": 0.00048239719157472415, "loss": 3.3364, "theoretical_loss": 4.526351407101618, "tokens_seen": 148045824 }, { "epoch": 0.04, "learning_rate": 0.0004823871614844534, "loss": 3.0166, "theoretical_loss": 4.526089025993732, "tokens_seen": 148111360 }, { "epoch": 0.04, "learning_rate": 0.00048237713139418257, "loss": 3.263, "theoretical_loss": 4.525826793449008, "tokens_seen": 148176896 }, { "epoch": 0.04, "learning_rate": 0.00048236710130391175, "loss": 3.4591, "theoretical_loss": 4.525564709317678, "tokens_seen": 148242432 }, { "epoch": 0.04, "learning_rate": 0.00048235707121364093, "loss": 3.4337, "theoretical_loss": 4.525302773450187, "tokens_seen": 148307968 }, { "epoch": 0.04, "learning_rate": 0.0004823470411233701, "loss": 3.3084, "theoretical_loss": 4.525040985697203, "tokens_seen": 148373504 }, { "epoch": 0.04, "learning_rate": 0.0004823370110330993, "loss": 3.5317, "theoretical_loss": 4.524779345909604, "tokens_seen": 148439040 }, { "epoch": 0.05, "learning_rate": 0.00048232698094282853, "loss": 3.4339, "theoretical_loss": 4.524517853938489, "tokens_seen": 148504576 }, { "epoch": 0.05, "learning_rate": 0.00048231695085255766, "loss": 3.1791, "theoretical_loss": 4.524256509635169, "tokens_seen": 148570112 }, { "epoch": 0.05, "learning_rate": 0.0004823069207622869, "loss": 3.1341, "theoretical_loss": 4.523995312851174, "tokens_seen": 148635648 }, { "epoch": 0.05, "learning_rate": 0.000482296890672016, "loss": 3.2603, "theoretical_loss": 4.523734263438241, "tokens_seen": 148701184 }, { "epoch": 0.05, "learning_rate": 0.00048228686058174525, "loss": 3.2534, "theoretical_loss": 4.52347336124833, "tokens_seen": 148766720 }, { "epoch": 0.05, "learning_rate": 0.00048227683049147443, "loss": 3.4183, "theoretical_loss": 4.52321260613361, "tokens_seen": 148832256 }, { "epoch": 0.05, "learning_rate": 0.0004822668004012036, "loss": 3.3214, "theoretical_loss": 4.522951997946466, "tokens_seen": 148897792 }, { "epoch": 0.05, "learning_rate": 0.0004822567703109328, "loss": 3.3674, "theoretical_loss": 4.522691536539492, "tokens_seen": 148963328 }, { "epoch": 0.05, "learning_rate": 0.00048224674022066203, "loss": 3.2784, "theoretical_loss": 4.522431221765498, "tokens_seen": 149028864 }, { "epoch": 0.05, "objective/train/docs_used": 95869, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1724908351898193, "objective/train/theoretical_loss": 4.522171053477507, "objective/train/tokens_used": 169554400, "theoretical_loss": 4.522171053477507, "tokens_seen": 149094400 }, { "epoch": 0.05, "learning_rate": 0.00048223671013039116, "loss": 3.2823, "theoretical_loss": 4.522171053477507, "tokens_seen": 149094400 }, { "epoch": 0.05, "learning_rate": 0.0004822266800401204, "loss": 3.2289, "theoretical_loss": 4.5219110315287505, "tokens_seen": 149159936 }, { "epoch": 0.05, "learning_rate": 0.0004822166499498495, "loss": 3.3112, "theoretical_loss": 4.521651155772675, "tokens_seen": 149225472 }, { "epoch": 0.05, "learning_rate": 0.00048220661985957876, "loss": 3.5058, "theoretical_loss": 4.521391426062934, "tokens_seen": 149291008 }, { "epoch": 0.05, "learning_rate": 0.00048219658976930794, "loss": 3.5517, "theoretical_loss": 4.521131842253396, "tokens_seen": 149356544 }, { "epoch": 0.05, "learning_rate": 0.0004821865596790371, "loss": 3.4992, "theoretical_loss": 4.520872404198139, "tokens_seen": 149422080 }, { "epoch": 0.05, "learning_rate": 0.0004821765295887663, "loss": 3.4315, "theoretical_loss": 4.520613111751445, "tokens_seen": 149487616 }, { "epoch": 0.05, "learning_rate": 0.0004821664994984955, "loss": 3.3585, "theoretical_loss": 4.520353964767814, "tokens_seen": 149553152 }, { "epoch": 0.05, "learning_rate": 0.00048215646940822466, "loss": 3.3818, "theoretical_loss": 4.5200949631019505, "tokens_seen": 149618688 }, { "epoch": 0.05, "learning_rate": 0.0004821464393179539, "loss": 3.183, "theoretical_loss": 4.519836106608768, "tokens_seen": 149684224 }, { "epoch": 0.05, "learning_rate": 0.000482136409227683, "loss": 3.2419, "theoretical_loss": 4.519577395143388, "tokens_seen": 149749760 }, { "epoch": 0.05, "learning_rate": 0.00048212637913741226, "loss": 3.139, "theoretical_loss": 4.519318828561142, "tokens_seen": 149815296 }, { "epoch": 0.05, "learning_rate": 0.0004821163490471414, "loss": 3.1937, "theoretical_loss": 4.519060406717565, "tokens_seen": 149880832 }, { "epoch": 0.05, "learning_rate": 0.0004821063189568706, "loss": 3.1968, "theoretical_loss": 4.518802129468405, "tokens_seen": 149946368 }, { "epoch": 0.05, "learning_rate": 0.0004820962888665998, "loss": 3.0867, "theoretical_loss": 4.51854399666961, "tokens_seen": 150011904 }, { "epoch": 0.05, "learning_rate": 0.000482086258776329, "loss": 3.1493, "theoretical_loss": 4.518286008177341, "tokens_seen": 150077440 }, { "epoch": 0.05, "learning_rate": 0.00048207622868605816, "loss": 3.4273, "theoretical_loss": 4.51802816384796, "tokens_seen": 150142976 }, { "epoch": 0.05, "learning_rate": 0.0004820661985957874, "loss": 3.3718, "theoretical_loss": 4.517770463538038, "tokens_seen": 150208512 }, { "epoch": 0.05, "learning_rate": 0.00048205616850551653, "loss": 3.4009, "theoretical_loss": 4.517512907104347, "tokens_seen": 150274048 }, { "epoch": 0.05, "learning_rate": 0.00048204613841524576, "loss": 3.2267, "theoretical_loss": 4.517255494403868, "tokens_seen": 150339584 }, { "epoch": 0.05, "learning_rate": 0.0004820361083249749, "loss": 3.2091, "theoretical_loss": 4.516998225293785, "tokens_seen": 150405120 }, { "epoch": 0.05, "learning_rate": 0.0004820260782347041, "loss": 3.4881, "theoretical_loss": 4.516741099631485, "tokens_seen": 150470656 }, { "epoch": 0.05, "learning_rate": 0.0004820160481444333, "loss": 3.2292, "theoretical_loss": 4.51648411727456, "tokens_seen": 150536192 }, { "epoch": 0.05, "learning_rate": 0.0004820060180541625, "loss": 3.291, "theoretical_loss": 4.5162272780808035, "tokens_seen": 150601728 }, { "epoch": 0.05, "learning_rate": 0.00048199598796389167, "loss": 3.1278, "theoretical_loss": 4.515970581908216, "tokens_seen": 150667264 }, { "epoch": 0.05, "objective/train/docs_used": 96442, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.100712299346924, "objective/train/theoretical_loss": 4.515714028614996, "objective/train/tokens_used": 171192800, "theoretical_loss": 4.515714028614996, "tokens_seen": 150732800 }, { "epoch": 0.05, "learning_rate": 0.00048198595787362085, "loss": 3.182, "theoretical_loss": 4.515714028614996, "tokens_seen": 150732800 }, { "epoch": 0.05, "learning_rate": 0.0004819759277833501, "loss": 3.4292, "theoretical_loss": 4.515457618059546, "tokens_seen": 150798336 }, { "epoch": 0.05, "learning_rate": 0.00048196589769307927, "loss": 3.469, "theoretical_loss": 4.515201350100471, "tokens_seen": 150863872 }, { "epoch": 0.05, "learning_rate": 0.00048195586760280845, "loss": 3.0502, "theoretical_loss": 4.514945224596577, "tokens_seen": 150929408 }, { "epoch": 0.05, "learning_rate": 0.00048194583751253763, "loss": 3.0543, "theoretical_loss": 4.5146892414068684, "tokens_seen": 150994944 }, { "epoch": 0.05, "learning_rate": 0.00048193580742226686, "loss": 3.1962, "theoretical_loss": 4.514433400390554, "tokens_seen": 151060480 }, { "epoch": 0.05, "learning_rate": 0.000481925777331996, "loss": 3.2645, "theoretical_loss": 4.514177701407042, "tokens_seen": 151126016 }, { "epoch": 0.05, "learning_rate": 0.0004819157472417252, "loss": 3.411, "theoretical_loss": 4.51392214431594, "tokens_seen": 151191552 }, { "epoch": 0.05, "learning_rate": 0.00048190571715145435, "loss": 3.4078, "theoretical_loss": 4.513666728977054, "tokens_seen": 151257088 }, { "epoch": 0.05, "learning_rate": 0.0004818956870611836, "loss": 3.1651, "theoretical_loss": 4.51341145525039, "tokens_seen": 151322624 }, { "epoch": 0.05, "learning_rate": 0.00048188565697091277, "loss": 3.0852, "theoretical_loss": 4.513156322996155, "tokens_seen": 151388160 }, { "epoch": 0.05, "learning_rate": 0.00048187562688064195, "loss": 3.223, "theoretical_loss": 4.512901332074751, "tokens_seen": 151453696 }, { "epoch": 0.05, "learning_rate": 0.00048186559679037113, "loss": 3.2013, "theoretical_loss": 4.5126464823467805, "tokens_seen": 151519232 }, { "epoch": 0.05, "learning_rate": 0.0004818555667001003, "loss": 3.4013, "theoretical_loss": 4.512391773673042, "tokens_seen": 151584768 }, { "epoch": 0.05, "learning_rate": 0.0004818455366098295, "loss": 3.1806, "theoretical_loss": 4.5121372059145335, "tokens_seen": 151650304 }, { "epoch": 0.05, "learning_rate": 0.00048183550651955873, "loss": 3.0253, "theoretical_loss": 4.511882778932447, "tokens_seen": 151715840 }, { "epoch": 0.05, "learning_rate": 0.00048182547642928786, "loss": 3.437, "theoretical_loss": 4.511628492588174, "tokens_seen": 151781376 }, { "epoch": 0.05, "learning_rate": 0.0004818154463390171, "loss": 3.2479, "theoretical_loss": 4.5113743467433, "tokens_seen": 151846912 }, { "epoch": 0.05, "learning_rate": 0.0004818054162487462, "loss": 3.2266, "theoretical_loss": 4.511120341259608, "tokens_seen": 151912448 }, { "epoch": 0.05, "learning_rate": 0.00048179538615847545, "loss": 3.065, "theoretical_loss": 4.510866475999077, "tokens_seen": 151977984 }, { "epoch": 0.05, "learning_rate": 0.00048178535606820463, "loss": 3.1586, "theoretical_loss": 4.510612750823878, "tokens_seen": 152043520 }, { "epoch": 0.05, "learning_rate": 0.0004817753259779338, "loss": 2.9635, "theoretical_loss": 4.5103591655963795, "tokens_seen": 152109056 }, { "epoch": 0.05, "learning_rate": 0.000481765295887663, "loss": 3.2094, "theoretical_loss": 4.510105720179144, "tokens_seen": 152174592 }, { "epoch": 0.05, "learning_rate": 0.00048175526579739223, "loss": 3.1597, "theoretical_loss": 4.5098524144349295, "tokens_seen": 152240128 }, { "epoch": 0.05, "learning_rate": 0.00048174523570712136, "loss": 3.1867, "theoretical_loss": 4.509599248226683, "tokens_seen": 152305664 }, { "epoch": 0.05, "objective/train/docs_used": 97778, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6629927158355713, "objective/train/theoretical_loss": 4.509346221417552, "objective/train/tokens_used": 172831200, "theoretical_loss": 4.509346221417552, "tokens_seen": 152371200 }, { "epoch": 0.05, "learning_rate": 0.0004817352056168506, "loss": 3.2688, "theoretical_loss": 4.509346221417552, "tokens_seen": 152371200 }, { "epoch": 0.05, "learning_rate": 0.0004817251755265797, "loss": 3.4886, "theoretical_loss": 4.509093333870869, "tokens_seen": 152436736 }, { "epoch": 0.05, "learning_rate": 0.00048171514543630896, "loss": 3.2402, "theoretical_loss": 4.508840585450166, "tokens_seen": 152502272 }, { "epoch": 0.05, "learning_rate": 0.00048170511534603814, "loss": 3.3127, "theoretical_loss": 4.508587976019164, "tokens_seen": 152567808 }, { "epoch": 0.05, "learning_rate": 0.0004816950852557673, "loss": 3.2319, "theoretical_loss": 4.508335505441774, "tokens_seen": 152633344 }, { "epoch": 0.05, "learning_rate": 0.0004816850551654965, "loss": 3.3341, "theoretical_loss": 4.508083173582105, "tokens_seen": 152698880 }, { "epoch": 0.05, "learning_rate": 0.0004816750250752257, "loss": 3.0618, "theoretical_loss": 4.507830980304451, "tokens_seen": 152764416 }, { "epoch": 0.05, "learning_rate": 0.00048166499498495486, "loss": 3.4654, "theoretical_loss": 4.5075789254733, "tokens_seen": 152829952 }, { "epoch": 0.05, "learning_rate": 0.0004816549648946841, "loss": 2.7021, "theoretical_loss": 4.507327008953329, "tokens_seen": 152895488 }, { "epoch": 0.05, "learning_rate": 0.0004816449348044132, "loss": 3.5096, "theoretical_loss": 4.507075230609407, "tokens_seen": 152961024 }, { "epoch": 0.05, "learning_rate": 0.00048163490471414246, "loss": 3.1766, "theoretical_loss": 4.506823590306591, "tokens_seen": 153026560 }, { "epoch": 0.05, "learning_rate": 0.0004816248746238716, "loss": 3.4232, "theoretical_loss": 4.506572087910127, "tokens_seen": 153092096 }, { "epoch": 0.05, "learning_rate": 0.0004816148445336008, "loss": 3.3578, "theoretical_loss": 4.506320723285455, "tokens_seen": 153157632 }, { "epoch": 0.05, "learning_rate": 0.00048160481444333, "loss": 3.4206, "theoretical_loss": 4.506069496298198, "tokens_seen": 153223168 }, { "epoch": 0.05, "learning_rate": 0.0004815947843530592, "loss": 3.5393, "theoretical_loss": 4.5058184068141705, "tokens_seen": 153288704 }, { "epoch": 0.05, "learning_rate": 0.00048158475426278837, "loss": 2.9682, "theoretical_loss": 4.505567454699373, "tokens_seen": 153354240 }, { "epoch": 0.05, "learning_rate": 0.0004815747241725176, "loss": 3.3476, "theoretical_loss": 4.505316639819997, "tokens_seen": 153419776 }, { "epoch": 0.05, "learning_rate": 0.00048156469408224673, "loss": 3.4125, "theoretical_loss": 4.505065962042418, "tokens_seen": 153485312 }, { "epoch": 0.05, "learning_rate": 0.00048155466399197596, "loss": 3.0099, "theoretical_loss": 4.504815421233202, "tokens_seen": 153550848 }, { "epoch": 0.05, "learning_rate": 0.0004815446339017051, "loss": 3.2099, "theoretical_loss": 4.504565017259097, "tokens_seen": 153616384 }, { "epoch": 0.05, "learning_rate": 0.0004815346038114343, "loss": 3.1983, "theoretical_loss": 4.504314749987044, "tokens_seen": 153681920 }, { "epoch": 0.05, "learning_rate": 0.0004815245737211635, "loss": 3.4997, "theoretical_loss": 4.504064619284163, "tokens_seen": 153747456 }, { "epoch": 0.05, "learning_rate": 0.0004815145436308927, "loss": 3.3542, "theoretical_loss": 4.503814625017766, "tokens_seen": 153812992 }, { "epoch": 0.05, "learning_rate": 0.00048150451354062187, "loss": 3.4172, "theoretical_loss": 4.5035647670553445, "tokens_seen": 153878528 }, { "epoch": 0.05, "learning_rate": 0.00048149448345035105, "loss": 3.0587, "theoretical_loss": 4.503315045264581, "tokens_seen": 153944064 }, { "epoch": 0.05, "objective/train/docs_used": 98099, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.236271858215332, "objective/train/theoretical_loss": 4.503065459513339, "objective/train/tokens_used": 174469600, "theoretical_loss": 4.503065459513339, "tokens_seen": 154009600 }, { "epoch": 0.05, "learning_rate": 0.00048148445336008023, "loss": 3.1071, "theoretical_loss": 4.503065459513339, "tokens_seen": 154009600 }, { "epoch": 0.05, "learning_rate": 0.00048147442326980947, "loss": 3.4448, "theoretical_loss": 4.502816009669665, "tokens_seen": 154075136 }, { "epoch": 0.05, "learning_rate": 0.0004814643931795386, "loss": 3.301, "theoretical_loss": 4.502566695601795, "tokens_seen": 154140672 }, { "epoch": 0.05, "learning_rate": 0.00048145436308926783, "loss": 3.3052, "theoretical_loss": 4.502317517178142, "tokens_seen": 154206208 }, { "epoch": 0.05, "learning_rate": 0.00048144433299899696, "loss": 3.1486, "theoretical_loss": 4.502068474267309, "tokens_seen": 154271744 }, { "epoch": 0.05, "learning_rate": 0.0004814343029087262, "loss": 2.9085, "theoretical_loss": 4.501819566738076, "tokens_seen": 154337280 }, { "epoch": 0.05, "learning_rate": 0.00048142427281845537, "loss": 3.271, "theoretical_loss": 4.501570794459411, "tokens_seen": 154402816 }, { "epoch": 0.05, "learning_rate": 0.00048141424272818455, "loss": 3.4126, "theoretical_loss": 4.501322157300461, "tokens_seen": 154468352 }, { "epoch": 0.05, "learning_rate": 0.00048140421263791373, "loss": 3.3014, "theoretical_loss": 4.501073655130554, "tokens_seen": 154533888 }, { "epoch": 0.05, "learning_rate": 0.00048139418254764297, "loss": 3.3142, "theoretical_loss": 4.500825287819205, "tokens_seen": 154599424 }, { "epoch": 0.05, "learning_rate": 0.0004813841524573721, "loss": 3.3548, "theoretical_loss": 4.500577055236104, "tokens_seen": 154664960 }, { "epoch": 0.05, "learning_rate": 0.00048137412236710133, "loss": 3.179, "theoretical_loss": 4.500328957251128, "tokens_seen": 154730496 }, { "epoch": 0.05, "learning_rate": 0.00048136409227683046, "loss": 3.2137, "theoretical_loss": 4.500080993734329, "tokens_seen": 154796032 }, { "epoch": 0.05, "learning_rate": 0.0004813540621865597, "loss": 3.3226, "theoretical_loss": 4.499833164555944, "tokens_seen": 154861568 }, { "epoch": 0.05, "learning_rate": 0.0004813440320962889, "loss": 3.3445, "theoretical_loss": 4.499585469586387, "tokens_seen": 154927104 }, { "epoch": 0.05, "learning_rate": 0.00048133400200601806, "loss": 3.2798, "theoretical_loss": 4.499337908696255, "tokens_seen": 154992640 }, { "epoch": 0.05, "learning_rate": 0.00048132397191574724, "loss": 3.1576, "theoretical_loss": 4.499090481756321, "tokens_seen": 155058176 }, { "epoch": 0.05, "learning_rate": 0.0004813139418254764, "loss": 3.3429, "theoretical_loss": 4.498843188637538, "tokens_seen": 155123712 }, { "epoch": 0.05, "learning_rate": 0.0004813039117352056, "loss": 3.2675, "theoretical_loss": 4.498596029211041, "tokens_seen": 155189248 }, { "epoch": 0.05, "learning_rate": 0.00048129388164493483, "loss": 3.3317, "theoretical_loss": 4.498349003348137, "tokens_seen": 155254784 }, { "epoch": 0.05, "learning_rate": 0.00048128385155466396, "loss": 3.3357, "theoretical_loss": 4.4981021109203185, "tokens_seen": 155320320 }, { "epoch": 0.05, "learning_rate": 0.0004812738214643932, "loss": 3.4323, "theoretical_loss": 4.49785535179925, "tokens_seen": 155385856 }, { "epoch": 0.05, "learning_rate": 0.0004812637913741223, "loss": 3.3122, "theoretical_loss": 4.497608725856776, "tokens_seen": 155451392 }, { "epoch": 0.05, "learning_rate": 0.00048125376128385156, "loss": 3.2995, "theoretical_loss": 4.497362232964919, "tokens_seen": 155516928 }, { "epoch": 0.05, "learning_rate": 0.00048124373119358074, "loss": 3.3192, "theoretical_loss": 4.497115872995876, "tokens_seen": 155582464 }, { "epoch": 0.05, "objective/train/docs_used": 99496, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4420876502990723, "objective/train/theoretical_loss": 4.496869645822022, "objective/train/tokens_used": 176108000, "theoretical_loss": 4.496869645822022, "tokens_seen": 155648000 }, { "epoch": 0.05, "learning_rate": 0.0004812337011033099, "loss": 3.2141, "theoretical_loss": 4.496869645822022, "tokens_seen": 155648000 }, { "epoch": 0.05, "learning_rate": 0.00048122367101303916, "loss": 3.2491, "theoretical_loss": 4.496623551315908, "tokens_seen": 155713536 }, { "epoch": 0.05, "learning_rate": 0.00048121364092276834, "loss": 3.5781, "theoretical_loss": 4.496377589350261, "tokens_seen": 155779072 }, { "epoch": 0.05, "learning_rate": 0.0004812036108324975, "loss": 3.5251, "theoretical_loss": 4.496131759797984, "tokens_seen": 155844608 }, { "epoch": 0.05, "learning_rate": 0.0004811935807422267, "loss": 3.1931, "theoretical_loss": 4.495886062532153, "tokens_seen": 155910144 }, { "epoch": 0.05, "learning_rate": 0.0004811835506519559, "loss": 3.425, "theoretical_loss": 4.495640497426023, "tokens_seen": 155975680 }, { "epoch": 0.05, "learning_rate": 0.00048117352056168506, "loss": 3.1489, "theoretical_loss": 4.495395064353019, "tokens_seen": 156041216 }, { "epoch": 0.05, "learning_rate": 0.0004811634904714143, "loss": 3.3909, "theoretical_loss": 4.4951497631867445, "tokens_seen": 156106752 }, { "epoch": 0.05, "learning_rate": 0.0004811534603811434, "loss": 3.505, "theoretical_loss": 4.494904593800973, "tokens_seen": 156172288 }, { "epoch": 0.05, "learning_rate": 0.00048114343029087266, "loss": 3.4255, "theoretical_loss": 4.4946595560696565, "tokens_seen": 156237824 }, { "epoch": 0.05, "learning_rate": 0.0004811334002006018, "loss": 3.2254, "theoretical_loss": 4.494414649866915, "tokens_seen": 156303360 }, { "epoch": 0.05, "learning_rate": 0.000481123370110331, "loss": 3.285, "theoretical_loss": 4.494169875067046, "tokens_seen": 156368896 }, { "epoch": 0.05, "learning_rate": 0.0004811133400200602, "loss": 3.1795, "theoretical_loss": 4.493925231544516, "tokens_seen": 156434432 }, { "epoch": 0.05, "learning_rate": 0.0004811033099297894, "loss": 3.4117, "theoretical_loss": 4.493680719173968, "tokens_seen": 156499968 }, { "epoch": 0.05, "learning_rate": 0.00048109327983951857, "loss": 3.1854, "theoretical_loss": 4.4934363378302145, "tokens_seen": 156565504 }, { "epoch": 0.05, "learning_rate": 0.0004810832497492478, "loss": 3.1177, "theoretical_loss": 4.493192087388239, "tokens_seen": 156631040 }, { "epoch": 0.05, "learning_rate": 0.00048107321965897693, "loss": 3.3501, "theoretical_loss": 4.4929479677232, "tokens_seen": 156696576 }, { "epoch": 0.05, "learning_rate": 0.00048106318956870616, "loss": 3.3047, "theoretical_loss": 4.4927039787104235, "tokens_seen": 156762112 }, { "epoch": 0.05, "learning_rate": 0.0004810531594784353, "loss": 3.3947, "theoretical_loss": 4.4924601202254095, "tokens_seen": 156827648 }, { "epoch": 0.05, "learning_rate": 0.0004810431293881645, "loss": 3.217, "theoretical_loss": 4.492216392143826, "tokens_seen": 156893184 }, { "epoch": 0.05, "learning_rate": 0.0004810330992978937, "loss": 3.3287, "theoretical_loss": 4.491972794341514, "tokens_seen": 156958720 }, { "epoch": 0.05, "learning_rate": 0.0004810230692076229, "loss": 3.4629, "theoretical_loss": 4.49172932669448, "tokens_seen": 157024256 }, { "epoch": 0.05, "learning_rate": 0.00048101303911735207, "loss": 3.2084, "theoretical_loss": 4.491485989078906, "tokens_seen": 157089792 }, { "epoch": 0.05, "learning_rate": 0.00048100300902708125, "loss": 3.4562, "theoretical_loss": 4.491242781371138, "tokens_seen": 157155328 }, { "epoch": 0.05, "learning_rate": 0.00048099297893681043, "loss": 3.2665, "theoretical_loss": 4.490999703447697, "tokens_seen": 157220864 }, { "epoch": 0.05, "objective/train/docs_used": 100000, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1658103466033936, "objective/train/theoretical_loss": 4.4907567551852665, "objective/train/tokens_used": 177746400, "theoretical_loss": 4.4907567551852665, "tokens_seen": 157286400 }, { "epoch": 0.05, "learning_rate": 0.00048098294884653967, "loss": 3.3429, "theoretical_loss": 4.4907567551852665, "tokens_seen": 157286400 }, { "epoch": 0.05, "learning_rate": 0.0004809729187562688, "loss": 3.416, "theoretical_loss": 4.490513936460702, "tokens_seen": 157351936 }, { "epoch": 0.05, "learning_rate": 0.00048096288866599803, "loss": 3.1769, "theoretical_loss": 4.490271247151027, "tokens_seen": 157417472 }, { "epoch": 0.05, "learning_rate": 0.00048095285857572716, "loss": 3.4918, "theoretical_loss": 4.490028687133432, "tokens_seen": 157483008 }, { "epoch": 0.05, "learning_rate": 0.0004809428284854564, "loss": 3.1271, "theoretical_loss": 4.489786256285276, "tokens_seen": 157548544 }, { "epoch": 0.05, "learning_rate": 0.00048093279839518557, "loss": 3.3183, "theoretical_loss": 4.489543954484084, "tokens_seen": 157614080 }, { "epoch": 0.05, "learning_rate": 0.00048092276830491475, "loss": 3.176, "theoretical_loss": 4.489301781607551, "tokens_seen": 157679616 }, { "epoch": 0.05, "learning_rate": 0.00048091273821464393, "loss": 3.2113, "theoretical_loss": 4.489059737533534, "tokens_seen": 157745152 }, { "epoch": 0.05, "learning_rate": 0.00048090270812437317, "loss": 3.2051, "theoretical_loss": 4.48881782214006, "tokens_seen": 157810688 }, { "epoch": 0.05, "learning_rate": 0.0004808926780341023, "loss": 3.2277, "theoretical_loss": 4.48857603530532, "tokens_seen": 157876224 }, { "epoch": 0.05, "learning_rate": 0.00048088264794383153, "loss": 3.3081, "theoretical_loss": 4.488334376907673, "tokens_seen": 157941760 }, { "epoch": 0.05, "learning_rate": 0.00048087261785356066, "loss": 3.4009, "theoretical_loss": 4.4880928468256425, "tokens_seen": 158007296 }, { "epoch": 0.05, "learning_rate": 0.0004808625877632899, "loss": 3.2696, "theoretical_loss": 4.487851444937916, "tokens_seen": 158072832 }, { "epoch": 0.05, "learning_rate": 0.0004808525576730191, "loss": 3.3747, "theoretical_loss": 4.487610171123347, "tokens_seen": 158138368 }, { "epoch": 0.05, "learning_rate": 0.00048084252758274826, "loss": 3.2603, "theoretical_loss": 4.487369025260954, "tokens_seen": 158203904 }, { "epoch": 0.05, "learning_rate": 0.00048083249749247744, "loss": 3.3168, "theoretical_loss": 4.48712800722992, "tokens_seen": 158269440 }, { "epoch": 0.05, "learning_rate": 0.0004808224674022066, "loss": 3.0748, "theoretical_loss": 4.48688711690959, "tokens_seen": 158334976 }, { "epoch": 0.05, "learning_rate": 0.0004808124373119358, "loss": 3.2276, "theoretical_loss": 4.486646354179475, "tokens_seen": 158400512 }, { "epoch": 0.05, "learning_rate": 0.00048080240722166503, "loss": 3.3016, "theoretical_loss": 4.48640571891925, "tokens_seen": 158466048 }, { "epoch": 0.05, "learning_rate": 0.00048079237713139416, "loss": 3.2242, "theoretical_loss": 4.48616521100875, "tokens_seen": 158531584 }, { "epoch": 0.05, "learning_rate": 0.0004807823470411234, "loss": 3.5084, "theoretical_loss": 4.485924830327974, "tokens_seen": 158597120 }, { "epoch": 0.05, "learning_rate": 0.0004807723169508525, "loss": 3.2451, "theoretical_loss": 4.485684576757087, "tokens_seen": 158662656 }, { "epoch": 0.05, "learning_rate": 0.00048076228686058176, "loss": 3.5179, "theoretical_loss": 4.485444450176413, "tokens_seen": 158728192 }, { "epoch": 0.05, "learning_rate": 0.00048075225677031094, "loss": 3.4267, "theoretical_loss": 4.485204450466437, "tokens_seen": 158793728 }, { "epoch": 0.05, "learning_rate": 0.0004807422266800401, "loss": 3.4344, "theoretical_loss": 4.484964577507808, "tokens_seen": 158859264 }, { "epoch": 0.05, "objective/train/docs_used": 101476, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1423563957214355, "objective/train/theoretical_loss": 4.484724831181337, "objective/train/tokens_used": 179384800, "theoretical_loss": 4.484724831181337, "tokens_seen": 158924800 }, { "epoch": 0.05, "learning_rate": 0.0004807321965897693, "loss": 3.3062, "theoretical_loss": 4.484724831181337, "tokens_seen": 158924800 }, { "epoch": 0.05, "learning_rate": 0.00048072216649949854, "loss": 3.222, "theoretical_loss": 4.4844852113679945, "tokens_seen": 158990336 }, { "epoch": 0.05, "learning_rate": 0.00048071213640922766, "loss": 3.2831, "theoretical_loss": 4.484245717948913, "tokens_seen": 159055872 }, { "epoch": 0.05, "learning_rate": 0.0004807021063189569, "loss": 3.2262, "theoretical_loss": 4.484006350805385, "tokens_seen": 159121408 }, { "epoch": 0.05, "learning_rate": 0.000480692076228686, "loss": 3.3261, "theoretical_loss": 4.483767109818862, "tokens_seen": 159186944 }, { "epoch": 0.05, "learning_rate": 0.00048068204613841526, "loss": 3.483, "theoretical_loss": 4.483527994870958, "tokens_seen": 159252480 }, { "epoch": 0.05, "learning_rate": 0.00048067201604814444, "loss": 3.2439, "theoretical_loss": 4.483289005843445, "tokens_seen": 159318016 }, { "epoch": 0.05, "learning_rate": 0.0004806619859578736, "loss": 3.4449, "theoretical_loss": 4.483050142618255, "tokens_seen": 159383552 }, { "epoch": 0.05, "learning_rate": 0.0004806519558676028, "loss": 3.1009, "theoretical_loss": 4.482811405077482, "tokens_seen": 159449088 }, { "epoch": 0.05, "learning_rate": 0.000480641925777332, "loss": 3.2611, "theoretical_loss": 4.482572793103373, "tokens_seen": 159514624 }, { "epoch": 0.05, "learning_rate": 0.00048063189568706117, "loss": 3.3544, "theoretical_loss": 4.482334306578339, "tokens_seen": 159580160 }, { "epoch": 0.05, "learning_rate": 0.0004806218655967904, "loss": 3.1545, "theoretical_loss": 4.482095945384946, "tokens_seen": 159645696 }, { "epoch": 0.05, "learning_rate": 0.00048061183550651953, "loss": 3.2616, "theoretical_loss": 4.481857709405919, "tokens_seen": 159711232 }, { "epoch": 0.05, "learning_rate": 0.00048060180541624877, "loss": 3.3525, "theoretical_loss": 4.4816195985241425, "tokens_seen": 159776768 }, { "epoch": 0.05, "learning_rate": 0.00048059177532597795, "loss": 3.1274, "theoretical_loss": 4.481381612622657, "tokens_seen": 159842304 }, { "epoch": 0.05, "learning_rate": 0.00048058174523570713, "loss": 3.5142, "theoretical_loss": 4.481143751584659, "tokens_seen": 159907840 }, { "epoch": 0.05, "learning_rate": 0.0004805717151454363, "loss": 3.4438, "theoretical_loss": 4.480906015293505, "tokens_seen": 159973376 }, { "epoch": 0.05, "learning_rate": 0.0004805616850551655, "loss": 3.3428, "theoretical_loss": 4.480668403632706, "tokens_seen": 160038912 }, { "epoch": 0.05, "learning_rate": 0.00048055165496489467, "loss": 3.4011, "theoretical_loss": 4.480430916485929, "tokens_seen": 160104448 }, { "epoch": 0.05, "learning_rate": 0.0004805416248746239, "loss": 3.1368, "theoretical_loss": 4.480193553736999, "tokens_seen": 160169984 }, { "epoch": 0.05, "learning_rate": 0.00048053159478435303, "loss": 3.1091, "theoretical_loss": 4.479956315269897, "tokens_seen": 160235520 }, { "epoch": 0.05, "learning_rate": 0.00048052156469408227, "loss": 3.4397, "theoretical_loss": 4.479719200968757, "tokens_seen": 160301056 }, { "epoch": 0.05, "learning_rate": 0.0004805115346038114, "loss": 3.4909, "theoretical_loss": 4.479482210717871, "tokens_seen": 160366592 }, { "epoch": 0.05, "learning_rate": 0.00048050150451354063, "loss": 3.3116, "theoretical_loss": 4.479245344401685, "tokens_seen": 160432128 }, { "epoch": 0.05, "learning_rate": 0.0004804914744232698, "loss": 3.2769, "theoretical_loss": 4.479008601904798, "tokens_seen": 160497664 }, { "epoch": 0.05, "objective/train/docs_used": 102054, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4178307056427, "objective/train/theoretical_loss": 4.478771983111967, "objective/train/tokens_used": 181023200, "theoretical_loss": 4.478771983111967, "tokens_seen": 160563200 }, { "epoch": 0.05, "learning_rate": 0.000480481444332999, "loss": 3.467, "theoretical_loss": 4.478771983111967, "tokens_seen": 160563200 }, { "epoch": 0.05, "learning_rate": 0.00048047141424272823, "loss": 3.4466, "theoretical_loss": 4.478535487908101, "tokens_seen": 160628736 }, { "epoch": 0.05, "learning_rate": 0.00048046138415245736, "loss": 3.3249, "theoretical_loss": 4.478299116178265, "tokens_seen": 160694272 }, { "epoch": 0.05, "learning_rate": 0.0004804513540621866, "loss": 3.2936, "theoretical_loss": 4.478062867807674, "tokens_seen": 160759808 }, { "epoch": 0.05, "learning_rate": 0.00048044132397191577, "loss": 3.466, "theoretical_loss": 4.4778267426817, "tokens_seen": 160825344 }, { "epoch": 0.05, "learning_rate": 0.00048043129388164495, "loss": 3.4092, "theoretical_loss": 4.477590740685867, "tokens_seen": 160890880 }, { "epoch": 0.05, "learning_rate": 0.00048042126379137413, "loss": 3.2282, "theoretical_loss": 4.47735486170585, "tokens_seen": 160956416 }, { "epoch": 0.05, "learning_rate": 0.00048041123370110337, "loss": 3.088, "theoretical_loss": 4.47711910562748, "tokens_seen": 161021952 }, { "epoch": 0.05, "learning_rate": 0.0004804012036108325, "loss": 3.1203, "theoretical_loss": 4.4768834723367394, "tokens_seen": 161087488 }, { "epoch": 0.05, "learning_rate": 0.00048039117352056173, "loss": 3.5309, "theoretical_loss": 4.4766479617197605, "tokens_seen": 161153024 }, { "epoch": 0.05, "learning_rate": 0.00048038114343029086, "loss": 3.1474, "theoretical_loss": 4.476412573662829, "tokens_seen": 161218560 }, { "epoch": 0.05, "learning_rate": 0.0004803711133400201, "loss": 3.1828, "theoretical_loss": 4.4761773080523835, "tokens_seen": 161284096 }, { "epoch": 0.05, "learning_rate": 0.0004803610832497493, "loss": 3.3428, "theoretical_loss": 4.475942164775013, "tokens_seen": 161349632 }, { "epoch": 0.05, "learning_rate": 0.00048035105315947846, "loss": 3.0744, "theoretical_loss": 4.475707143717455, "tokens_seen": 161415168 }, { "epoch": 0.05, "learning_rate": 0.00048034102306920764, "loss": 3.1276, "theoretical_loss": 4.475472244766601, "tokens_seen": 161480704 }, { "epoch": 0.05, "learning_rate": 0.0004803309929789368, "loss": 3.0242, "theoretical_loss": 4.475237467809492, "tokens_seen": 161546240 }, { "epoch": 0.05, "learning_rate": 0.000480320962888666, "loss": 3.4158, "theoretical_loss": 4.47500281273332, "tokens_seen": 161611776 }, { "epoch": 0.05, "learning_rate": 0.00048031093279839523, "loss": 3.2473, "theoretical_loss": 4.474768279425424, "tokens_seen": 161677312 }, { "epoch": 0.05, "learning_rate": 0.00048030090270812436, "loss": 3.1312, "theoretical_loss": 4.474533867773299, "tokens_seen": 161742848 }, { "epoch": 0.05, "learning_rate": 0.0004802908726178536, "loss": 3.3392, "theoretical_loss": 4.474299577664581, "tokens_seen": 161808384 }, { "epoch": 0.05, "learning_rate": 0.0004802808425275827, "loss": 3.3211, "theoretical_loss": 4.474065408987063, "tokens_seen": 161873920 }, { "epoch": 0.05, "learning_rate": 0.00048027081243731196, "loss": 3.3298, "theoretical_loss": 4.473831361628682, "tokens_seen": 161939456 }, { "epoch": 0.05, "learning_rate": 0.00048026078234704114, "loss": 3.3306, "theoretical_loss": 4.473597435477526, "tokens_seen": 162004992 }, { "epoch": 0.05, "learning_rate": 0.0004802507522567703, "loss": 3.1812, "theoretical_loss": 4.473363630421831, "tokens_seen": 162070528 }, { "epoch": 0.05, "learning_rate": 0.0004802407221664995, "loss": 3.1926, "theoretical_loss": 4.473129946349982, "tokens_seen": 162136064 }, { "epoch": 0.05, "objective/train/docs_used": 102718, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.263267993927002, "objective/train/theoretical_loss": 4.472896383150508, "objective/train/tokens_used": 182661600, "theoretical_loss": 4.472896383150508, "tokens_seen": 162201600 }, { "epoch": 0.05, "learning_rate": 0.00048023069207622874, "loss": 3.1614, "theoretical_loss": 4.472896383150508, "tokens_seen": 162201600 }, { "epoch": 0.05, "learning_rate": 0.00048022066198595786, "loss": 3.4761, "theoretical_loss": 4.472662940712091, "tokens_seen": 162267136 }, { "epoch": 0.05, "learning_rate": 0.0004802106318956871, "loss": 3.1622, "theoretical_loss": 4.472429618923558, "tokens_seen": 162332672 }, { "epoch": 0.05, "learning_rate": 0.00048020060180541623, "loss": 3.2604, "theoretical_loss": 4.472196417673883, "tokens_seen": 162398208 }, { "epoch": 0.05, "learning_rate": 0.00048019057171514546, "loss": 3.373, "theoretical_loss": 4.471963336852187, "tokens_seen": 162463744 }, { "epoch": 0.05, "learning_rate": 0.00048018054162487464, "loss": 3.1342, "theoretical_loss": 4.471730376347738, "tokens_seen": 162529280 }, { "epoch": 0.05, "learning_rate": 0.0004801705115346038, "loss": 3.1844, "theoretical_loss": 4.4714975360499505, "tokens_seen": 162594816 }, { "epoch": 0.05, "learning_rate": 0.000480160481444333, "loss": 3.4885, "theoretical_loss": 4.471264815848384, "tokens_seen": 162660352 }, { "epoch": 0.05, "learning_rate": 0.0004801504513540622, "loss": 3.2073, "theoretical_loss": 4.471032215632746, "tokens_seen": 162725888 }, { "epoch": 0.05, "learning_rate": 0.00048014042126379137, "loss": 3.1441, "theoretical_loss": 4.470799735292889, "tokens_seen": 162791424 }, { "epoch": 0.05, "learning_rate": 0.0004801303911735206, "loss": 3.3775, "theoretical_loss": 4.470567374718808, "tokens_seen": 162856960 }, { "epoch": 0.05, "learning_rate": 0.00048012036108324973, "loss": 3.4188, "theoretical_loss": 4.470335133800649, "tokens_seen": 162922496 }, { "epoch": 0.05, "learning_rate": 0.00048011033099297897, "loss": 3.1071, "theoretical_loss": 4.470103012428696, "tokens_seen": 162988032 }, { "epoch": 0.05, "learning_rate": 0.00048010030090270815, "loss": 3.3467, "theoretical_loss": 4.469871010493383, "tokens_seen": 163053568 }, { "epoch": 0.05, "learning_rate": 0.00048009027081243733, "loss": 3.3758, "theoretical_loss": 4.469639127885287, "tokens_seen": 163119104 }, { "epoch": 0.05, "learning_rate": 0.0004800802407221665, "loss": 3.1418, "theoretical_loss": 4.4694073644951295, "tokens_seen": 163184640 }, { "epoch": 0.05, "learning_rate": 0.0004800702106318957, "loss": 3.2458, "theoretical_loss": 4.469175720213771, "tokens_seen": 163250176 }, { "epoch": 0.05, "learning_rate": 0.00048006018054162487, "loss": 3.3036, "theoretical_loss": 4.468944194932225, "tokens_seen": 163315712 }, { "epoch": 0.05, "learning_rate": 0.0004800501504513541, "loss": 3.2241, "theoretical_loss": 4.468712788541639, "tokens_seen": 163381248 }, { "epoch": 0.05, "learning_rate": 0.00048004012036108323, "loss": 3.2485, "theoretical_loss": 4.46848150093331, "tokens_seen": 163446784 }, { "epoch": 0.05, "learning_rate": 0.00048003009027081247, "loss": 3.286, "theoretical_loss": 4.468250331998676, "tokens_seen": 163512320 }, { "epoch": 0.05, "learning_rate": 0.0004800200601805416, "loss": 3.0841, "theoretical_loss": 4.468019281629316, "tokens_seen": 163577856 }, { "epoch": 0.05, "learning_rate": 0.00048001003009027083, "loss": 3.1869, "theoretical_loss": 4.467788349716955, "tokens_seen": 163643392 }, { "epoch": 0.05, "learning_rate": 0.00048, "loss": 3.2902, "theoretical_loss": 4.467557536153457, "tokens_seen": 163708928 }, { "epoch": 0.05, "learning_rate": 0.0004799899699097292, "loss": 3.179, "theoretical_loss": 4.467326840830829, "tokens_seen": 163774464 }, { "debugging/Self-BLEU-5": 0.47376287031010694, "debugging/distinct-1-grams": 0.7718165351312889, "debugging/distinct-2-grams": 0.9624982234606337, "debugging/entropy-1-grams": 5.847463360327344, "debugging/entropy-2-grams": 6.816946097439353, "debugging/length": 477.7857142857143, "debugging/num_segments": 14, "epoch": 0.05, "objective/train/docs_used": 104069, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2056386470794678, "objective/train/theoretical_loss": 4.467096263641219, "objective/train/tokens_used": 184300000, "theoretical_loss": 4.467096263641219, "tokens_seen": 163840000 }, { "epoch": 0.05, "learning_rate": 0.0004799799398194584, "loss": 3.24, "theoretical_loss": 4.467096263641219, "tokens_seen": 163840000 }, { "epoch": 0.05, "learning_rate": 0.00047996990972918756, "loss": 3.1111, "theoretical_loss": 4.466865804476919, "tokens_seen": 163905536 }, { "epoch": 0.05, "learning_rate": 0.00047995987963891674, "loss": 3.1284, "theoretical_loss": 4.466635463230359, "tokens_seen": 163971072 }, { "epoch": 0.05, "learning_rate": 0.00047994984954864597, "loss": 3.3292, "theoretical_loss": 4.466405239794113, "tokens_seen": 164036608 }, { "epoch": 0.05, "learning_rate": 0.0004799398194583751, "loss": 3.3509, "theoretical_loss": 4.466175134060894, "tokens_seen": 164102144 }, { "epoch": 0.05, "learning_rate": 0.00047992978936810433, "loss": 3.4643, "theoretical_loss": 4.465945145923554, "tokens_seen": 164167680 }, { "epoch": 0.05, "learning_rate": 0.0004799197592778335, "loss": 3.0926, "theoretical_loss": 4.4657152752750875, "tokens_seen": 164233216 }, { "epoch": 0.05, "learning_rate": 0.0004799097291875627, "loss": 3.2696, "theoretical_loss": 4.465485522008629, "tokens_seen": 164298752 }, { "epoch": 0.05, "learning_rate": 0.0004798996990972919, "loss": 3.158, "theoretical_loss": 4.465255886017452, "tokens_seen": 164364288 }, { "epoch": 0.05, "learning_rate": 0.00047988966900702106, "loss": 3.3011, "theoretical_loss": 4.465026367194971, "tokens_seen": 164429824 }, { "epoch": 0.05, "learning_rate": 0.00047987963891675024, "loss": 3.0533, "theoretical_loss": 4.464796965434738, "tokens_seen": 164495360 }, { "epoch": 0.05, "learning_rate": 0.0004798696088264795, "loss": 3.1821, "theoretical_loss": 4.464567680630443, "tokens_seen": 164560896 }, { "epoch": 0.05, "learning_rate": 0.0004798595787362086, "loss": 3.2458, "theoretical_loss": 4.464338512675919, "tokens_seen": 164626432 }, { "epoch": 0.05, "learning_rate": 0.00047984954864593784, "loss": 3.3066, "theoretical_loss": 4.464109461465133, "tokens_seen": 164691968 }, { "epoch": 0.05, "learning_rate": 0.00047983951855566696, "loss": 3.1252, "theoretical_loss": 4.4638805268921935, "tokens_seen": 164757504 }, { "epoch": 0.05, "learning_rate": 0.0004798294884653962, "loss": 3.1787, "theoretical_loss": 4.463651708851346, "tokens_seen": 164823040 }, { "epoch": 0.05, "learning_rate": 0.0004798194583751254, "loss": 3.209, "theoretical_loss": 4.463423007236974, "tokens_seen": 164888576 }, { "epoch": 0.05, "learning_rate": 0.00047980942828485456, "loss": 3.1093, "theoretical_loss": 4.4631944219436, "tokens_seen": 164954112 }, { "epoch": 0.05, "learning_rate": 0.00047979939819458374, "loss": 3.1704, "theoretical_loss": 4.462965952865879, "tokens_seen": 165019648 }, { "epoch": 0.05, "learning_rate": 0.0004797893681043129, "loss": 3.3571, "theoretical_loss": 4.46273759989861, "tokens_seen": 165085184 }, { "epoch": 0.05, "learning_rate": 0.0004797793380140421, "loss": 3.3585, "theoretical_loss": 4.462509362936723, "tokens_seen": 165150720 }, { "epoch": 0.05, "learning_rate": 0.00047976930792377134, "loss": 3.4484, "theoretical_loss": 4.46228124187529, "tokens_seen": 165216256 }, { "epoch": 0.05, "learning_rate": 0.00047975927783350047, "loss": 3.2982, "theoretical_loss": 4.462053236609516, "tokens_seen": 165281792 }, { "epoch": 0.05, "learning_rate": 0.0004797492477432297, "loss": 3.2709, "theoretical_loss": 4.461825347034742, "tokens_seen": 165347328 }, { "epoch": 0.05, "learning_rate": 0.0004797392176529589, "loss": 3.5718, "theoretical_loss": 4.461597573046449, "tokens_seen": 165412864 }, { "epoch": 0.05, "objective/train/docs_used": 105300, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3106820583343506, "objective/train/theoretical_loss": 4.461369914540247, "objective/train/tokens_used": 185938400, "theoretical_loss": 4.461369914540247, "tokens_seen": 165478400 }, { "epoch": 0.05, "learning_rate": 0.00047972918756268807, "loss": 3.3645, "theoretical_loss": 4.461369914540247, "tokens_seen": 165478400 }, { "epoch": 0.05, "learning_rate": 0.0004797191574724173, "loss": 3.322, "theoretical_loss": 4.4611423714118885, "tokens_seen": 165543936 }, { "epoch": 0.05, "learning_rate": 0.00047970912738214643, "loss": 3.3285, "theoretical_loss": 4.460914943557256, "tokens_seen": 165609472 }, { "epoch": 0.05, "learning_rate": 0.00047969909729187566, "loss": 3.2845, "theoretical_loss": 4.460687630872371, "tokens_seen": 165675008 }, { "epoch": 0.05, "learning_rate": 0.00047968906720160484, "loss": 3.2704, "theoretical_loss": 4.46046043325339, "tokens_seen": 165740544 }, { "epoch": 0.05, "learning_rate": 0.000479679037111334, "loss": 3.5378, "theoretical_loss": 4.460233350596599, "tokens_seen": 165806080 }, { "epoch": 0.05, "learning_rate": 0.0004796690070210632, "loss": 3.2292, "theoretical_loss": 4.460006382798425, "tokens_seen": 165871616 }, { "epoch": 0.05, "learning_rate": 0.0004796589769307924, "loss": 3.4241, "theoretical_loss": 4.459779529755423, "tokens_seen": 165937152 }, { "epoch": 0.05, "learning_rate": 0.00047964894684052157, "loss": 3.3032, "theoretical_loss": 4.459552791364288, "tokens_seen": 166002688 }, { "epoch": 0.05, "learning_rate": 0.0004796389167502508, "loss": 3.0771, "theoretical_loss": 4.459326167521844, "tokens_seen": 166068224 }, { "epoch": 0.05, "learning_rate": 0.00047962888665997993, "loss": 3.6786, "theoretical_loss": 4.4590996581250515, "tokens_seen": 166133760 }, { "epoch": 0.05, "learning_rate": 0.00047961885656970917, "loss": 3.4545, "theoretical_loss": 4.458873263071002, "tokens_seen": 166199296 }, { "epoch": 0.05, "learning_rate": 0.00047960882647943835, "loss": 3.228, "theoretical_loss": 4.458646982256921, "tokens_seen": 166264832 }, { "epoch": 0.05, "learning_rate": 0.00047959879638916753, "loss": 3.2841, "theoretical_loss": 4.458420815580169, "tokens_seen": 166330368 }, { "epoch": 0.05, "learning_rate": 0.0004795887662988967, "loss": 2.9158, "theoretical_loss": 4.458194762938234, "tokens_seen": 166395904 }, { "epoch": 0.05, "learning_rate": 0.0004795787362086259, "loss": 3.1962, "theoretical_loss": 4.457968824228743, "tokens_seen": 166461440 }, { "epoch": 0.05, "learning_rate": 0.00047956870611835507, "loss": 3.3127, "theoretical_loss": 4.457742999349449, "tokens_seen": 166526976 }, { "epoch": 0.05, "learning_rate": 0.0004795586760280843, "loss": 3.1228, "theoretical_loss": 4.4575172881982414, "tokens_seen": 166592512 }, { "epoch": 0.05, "learning_rate": 0.00047954864593781343, "loss": 3.3468, "theoretical_loss": 4.457291690673139, "tokens_seen": 166658048 }, { "epoch": 0.05, "learning_rate": 0.00047953861584754267, "loss": 3.1293, "theoretical_loss": 4.457066206672291, "tokens_seen": 166723584 }, { "epoch": 0.05, "learning_rate": 0.0004795285857572718, "loss": 3.4924, "theoretical_loss": 4.456840836093983, "tokens_seen": 166789120 }, { "epoch": 0.05, "learning_rate": 0.00047951855566700103, "loss": 3.3401, "theoretical_loss": 4.456615578836625, "tokens_seen": 166854656 }, { "epoch": 0.05, "learning_rate": 0.0004795085255767302, "loss": 3.3277, "theoretical_loss": 4.456390434798762, "tokens_seen": 166920192 }, { "epoch": 0.05, "learning_rate": 0.0004794984954864594, "loss": 3.1098, "theoretical_loss": 4.45616540387907, "tokens_seen": 166985728 }, { "epoch": 0.05, "learning_rate": 0.0004794884653961886, "loss": 3.3298, "theoretical_loss": 4.4559404859763525, "tokens_seen": 167051264 }, { "epoch": 0.05, "objective/train/docs_used": 106148, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4106695652008057, "objective/train/theoretical_loss": 4.455715680989545, "objective/train/tokens_used": 187576800, "theoretical_loss": 4.455715680989545, "tokens_seen": 167116800 }, { "epoch": 0.05, "learning_rate": 0.00047947843530591776, "loss": 2.9402, "theoretical_loss": 4.455715680989545, "tokens_seen": 167116800 }, { "epoch": 0.05, "learning_rate": 0.00047946840521564694, "loss": 3.432, "theoretical_loss": 4.455490988817713, "tokens_seen": 167182336 }, { "epoch": 0.05, "learning_rate": 0.00047945837512537617, "loss": 3.1718, "theoretical_loss": 4.4552664093600525, "tokens_seen": 167247872 }, { "epoch": 0.05, "learning_rate": 0.0004794483450351053, "loss": 3.1644, "theoretical_loss": 4.455041942515887, "tokens_seen": 167313408 }, { "epoch": 0.05, "learning_rate": 0.00047943831494483453, "loss": 3.1954, "theoretical_loss": 4.454817588184669, "tokens_seen": 167378944 }, { "epoch": 0.05, "learning_rate": 0.0004794282848545637, "loss": 3.2134, "theoretical_loss": 4.454593346265984, "tokens_seen": 167444480 }, { "epoch": 0.05, "learning_rate": 0.0004794182547642929, "loss": 3.4663, "theoretical_loss": 4.454369216659542, "tokens_seen": 167510016 }, { "epoch": 0.05, "learning_rate": 0.0004794082246740221, "loss": 3.3214, "theoretical_loss": 4.454145199265183, "tokens_seen": 167575552 }, { "epoch": 0.05, "learning_rate": 0.00047939819458375126, "loss": 3.2135, "theoretical_loss": 4.453921293982877, "tokens_seen": 167641088 }, { "epoch": 0.05, "learning_rate": 0.00047938816449348044, "loss": 3.3655, "theoretical_loss": 4.453697500712722, "tokens_seen": 167706624 }, { "epoch": 0.05, "learning_rate": 0.0004793781344032097, "loss": 3.202, "theoretical_loss": 4.453473819354942, "tokens_seen": 167772160 }, { "epoch": 0.05, "learning_rate": 0.0004793681043129388, "loss": 3.2527, "theoretical_loss": 4.453250249809889, "tokens_seen": 167837696 }, { "epoch": 0.05, "learning_rate": 0.00047935807422266804, "loss": 3.2188, "theoretical_loss": 4.453026791978045, "tokens_seen": 167903232 }, { "epoch": 0.05, "learning_rate": 0.00047934804413239716, "loss": 3.2968, "theoretical_loss": 4.4528034457600185, "tokens_seen": 167968768 }, { "epoch": 0.05, "learning_rate": 0.0004793380140421264, "loss": 3.133, "theoretical_loss": 4.452580211056542, "tokens_seen": 168034304 }, { "epoch": 0.05, "learning_rate": 0.0004793279839518556, "loss": 3.1103, "theoretical_loss": 4.452357087768481, "tokens_seen": 168099840 }, { "epoch": 0.05, "learning_rate": 0.00047931795386158476, "loss": 3.1566, "theoretical_loss": 4.45213407579682, "tokens_seen": 168165376 }, { "epoch": 0.05, "learning_rate": 0.00047930792377131394, "loss": 3.3208, "theoretical_loss": 4.451911175042679, "tokens_seen": 168230912 }, { "epoch": 0.05, "learning_rate": 0.0004792978936810431, "loss": 3.3578, "theoretical_loss": 4.451688385407296, "tokens_seen": 168296448 }, { "epoch": 0.05, "learning_rate": 0.0004792878635907723, "loss": 3.0301, "theoretical_loss": 4.451465706792041, "tokens_seen": 168361984 }, { "epoch": 0.05, "learning_rate": 0.00047927783350050154, "loss": 3.3597, "theoretical_loss": 4.4512431390984055, "tokens_seen": 168427520 }, { "epoch": 0.05, "learning_rate": 0.00047926780341023067, "loss": 3.2, "theoretical_loss": 4.451020682228011, "tokens_seen": 168493056 }, { "epoch": 0.05, "learning_rate": 0.0004792577733199599, "loss": 3.3903, "theoretical_loss": 4.450798336082601, "tokens_seen": 168558592 }, { "epoch": 0.05, "learning_rate": 0.0004792477432296891, "loss": 3.5916, "theoretical_loss": 4.450576100564046, "tokens_seen": 168624128 }, { "epoch": 0.05, "learning_rate": 0.00047923771313941827, "loss": 3.0493, "theoretical_loss": 4.450353975574341, "tokens_seen": 168689664 }, { "epoch": 0.05, "objective/train/docs_used": 106755, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.232574701309204, "objective/train/theoretical_loss": 4.450131961015606, "objective/train/tokens_used": 189215200, "theoretical_loss": 4.450131961015606, "tokens_seen": 168755200 }, { "epoch": 0.05, "learning_rate": 0.00047922768304914745, "loss": 3.3465, "theoretical_loss": 4.450131961015606, "tokens_seen": 168755200 }, { "epoch": 0.05, "learning_rate": 0.00047921765295887663, "loss": 3.4464, "theoretical_loss": 4.449910056790086, "tokens_seen": 168820736 }, { "epoch": 0.05, "learning_rate": 0.0004792076228686058, "loss": 3.2509, "theoretical_loss": 4.44968826280015, "tokens_seen": 168886272 }, { "epoch": 0.05, "learning_rate": 0.00047919759277833504, "loss": 3.4296, "theoretical_loss": 4.4494665789482895, "tokens_seen": 168951808 }, { "epoch": 0.05, "learning_rate": 0.00047918756268806417, "loss": 3.2661, "theoretical_loss": 4.449245005137125, "tokens_seen": 169017344 }, { "epoch": 0.05, "learning_rate": 0.0004791775325977934, "loss": 3.2495, "theoretical_loss": 4.449023541269395, "tokens_seen": 169082880 }, { "epoch": 0.05, "learning_rate": 0.00047916750250752253, "loss": 3.1478, "theoretical_loss": 4.448802187247966, "tokens_seen": 169148416 }, { "epoch": 0.05, "learning_rate": 0.00047915747241725177, "loss": 3.4552, "theoretical_loss": 4.448580942975825, "tokens_seen": 169213952 }, { "epoch": 0.05, "learning_rate": 0.00047914744232698095, "loss": 3.34, "theoretical_loss": 4.448359808356084, "tokens_seen": 169279488 }, { "epoch": 0.05, "learning_rate": 0.00047913741223671013, "loss": 3.2944, "theoretical_loss": 4.448138783291979, "tokens_seen": 169345024 }, { "epoch": 0.05, "learning_rate": 0.0004791273821464393, "loss": 3.2205, "theoretical_loss": 4.447917867686863, "tokens_seen": 169410560 }, { "epoch": 0.05, "learning_rate": 0.00047911735205616855, "loss": 3.1967, "theoretical_loss": 4.44769706144422, "tokens_seen": 169476096 }, { "epoch": 0.05, "learning_rate": 0.0004791073219658977, "loss": 3.0132, "theoretical_loss": 4.44747636446765, "tokens_seen": 169541632 }, { "epoch": 0.05, "learning_rate": 0.0004790972918756269, "loss": 3.4747, "theoretical_loss": 4.447255776660878, "tokens_seen": 169607168 }, { "epoch": 0.05, "learning_rate": 0.00047908726178535604, "loss": 3.3565, "theoretical_loss": 4.44703529792775, "tokens_seen": 169672704 }, { "epoch": 0.05, "learning_rate": 0.00047907723169508527, "loss": 3.0567, "theoretical_loss": 4.446814928172234, "tokens_seen": 169738240 }, { "epoch": 0.05, "learning_rate": 0.00047906720160481445, "loss": 3.1763, "theoretical_loss": 4.446594667298421, "tokens_seen": 169803776 }, { "epoch": 0.05, "learning_rate": 0.00047905717151454363, "loss": 3.3902, "theoretical_loss": 4.446374515210521, "tokens_seen": 169869312 }, { "epoch": 0.05, "learning_rate": 0.0004790471414242728, "loss": 3.4254, "theoretical_loss": 4.446154471812866, "tokens_seen": 169934848 }, { "epoch": 0.05, "learning_rate": 0.000479037111334002, "loss": 3.2961, "theoretical_loss": 4.445934537009911, "tokens_seen": 170000384 }, { "epoch": 0.05, "learning_rate": 0.0004790270812437312, "loss": 2.8793, "theoretical_loss": 4.445714710706228, "tokens_seen": 170065920 }, { "epoch": 0.05, "learning_rate": 0.0004790170511534604, "loss": 2.8962, "theoretical_loss": 4.445494992806513, "tokens_seen": 170131456 }, { "epoch": 0.05, "learning_rate": 0.00047900702106318954, "loss": 3.2095, "theoretical_loss": 4.44527538321558, "tokens_seen": 170196992 }, { "epoch": 0.05, "learning_rate": 0.0004789969909729188, "loss": 3.4441, "theoretical_loss": 4.445055881838365, "tokens_seen": 170262528 }, { "epoch": 0.05, "learning_rate": 0.0004789869608826479, "loss": 3.3936, "theoretical_loss": 4.444836488579924, "tokens_seen": 170328064 }, { "epoch": 0.05, "objective/train/docs_used": 107830, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1494503021240234, "objective/train/theoretical_loss": 4.44461720334543, "objective/train/tokens_used": 190853600, "theoretical_loss": 4.44461720334543, "tokens_seen": 170393600 }, { "epoch": 0.05, "learning_rate": 0.00047897693079237714, "loss": 3.2369, "theoretical_loss": 4.44461720334543, "tokens_seen": 170393600 }, { "epoch": 0.05, "learning_rate": 0.00047896690070210637, "loss": 3.3457, "theoretical_loss": 4.444398026040179, "tokens_seen": 170459136 }, { "epoch": 0.05, "learning_rate": 0.0004789568706118355, "loss": 3.3824, "theoretical_loss": 4.444178956569585, "tokens_seen": 170524672 }, { "epoch": 0.05, "learning_rate": 0.00047894684052156473, "loss": 3.4014, "theoretical_loss": 4.443959994839181, "tokens_seen": 170590208 }, { "epoch": 0.05, "learning_rate": 0.0004789368104312939, "loss": 3.3058, "theoretical_loss": 4.44374114075462, "tokens_seen": 170655744 }, { "epoch": 0.05, "learning_rate": 0.0004789267803410231, "loss": 3.1202, "theoretical_loss": 4.443522394221671, "tokens_seen": 170721280 }, { "epoch": 0.05, "learning_rate": 0.0004789167502507523, "loss": 3.3846, "theoretical_loss": 4.443303755146225, "tokens_seen": 170786816 }, { "epoch": 0.05, "learning_rate": 0.00047890672016048146, "loss": 3.2755, "theoretical_loss": 4.443085223434291, "tokens_seen": 170852352 }, { "epoch": 0.05, "learning_rate": 0.00047889669007021064, "loss": 3.0001, "theoretical_loss": 4.442866798991993, "tokens_seen": 170917888 }, { "epoch": 0.05, "learning_rate": 0.0004788866599799399, "loss": 3.5228, "theoretical_loss": 4.442648481725577, "tokens_seen": 170983424 }, { "epoch": 0.05, "learning_rate": 0.000478876629889669, "loss": 3.5152, "theoretical_loss": 4.442430271541404, "tokens_seen": 171048960 }, { "epoch": 0.05, "learning_rate": 0.00047886659979939824, "loss": 3.0459, "theoretical_loss": 4.442212168345956, "tokens_seen": 171114496 }, { "epoch": 0.05, "learning_rate": 0.00047885656970912736, "loss": 3.2609, "theoretical_loss": 4.4419941720458285, "tokens_seen": 171180032 }, { "epoch": 0.05, "learning_rate": 0.0004788465396188566, "loss": 3.1621, "theoretical_loss": 4.441776282547736, "tokens_seen": 171245568 }, { "epoch": 0.05, "learning_rate": 0.0004788365095285858, "loss": 3.2143, "theoretical_loss": 4.441558499758511, "tokens_seen": 171311104 }, { "epoch": 0.05, "learning_rate": 0.00047882647943831496, "loss": 3.3814, "theoretical_loss": 4.441340823585101, "tokens_seen": 171376640 }, { "epoch": 0.05, "learning_rate": 0.00047881644934804414, "loss": 3.4502, "theoretical_loss": 4.441123253934572, "tokens_seen": 171442176 }, { "epoch": 0.05, "learning_rate": 0.0004788064192577733, "loss": 3.2365, "theoretical_loss": 4.440905790714105, "tokens_seen": 171507712 }, { "epoch": 0.05, "learning_rate": 0.0004787963891675025, "loss": 2.9738, "theoretical_loss": 4.440688433830999, "tokens_seen": 171573248 }, { "epoch": 0.05, "learning_rate": 0.00047878635907723174, "loss": 3.2246, "theoretical_loss": 4.440471183192667, "tokens_seen": 171638784 }, { "epoch": 0.05, "learning_rate": 0.00047877632898696087, "loss": 3.3308, "theoretical_loss": 4.440254038706639, "tokens_seen": 171704320 }, { "epoch": 0.05, "learning_rate": 0.0004787662988966901, "loss": 3.1173, "theoretical_loss": 4.440037000280561, "tokens_seen": 171769856 }, { "epoch": 0.05, "learning_rate": 0.0004787562688064193, "loss": 3.2074, "theoretical_loss": 4.439820067822195, "tokens_seen": 171835392 }, { "epoch": 0.05, "learning_rate": 0.00047874623871614847, "loss": 3.115, "theoretical_loss": 4.439603241239416, "tokens_seen": 171900928 }, { "epoch": 0.05, "learning_rate": 0.00047873620862587765, "loss": 3.3423, "theoretical_loss": 4.439386520440218, "tokens_seen": 171966464 }, { "epoch": 0.05, "objective/train/docs_used": 108471, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.364562749862671, "objective/train/theoretical_loss": 4.439169905332706, "objective/train/tokens_used": 192492000, "theoretical_loss": 4.439169905332706, "tokens_seen": 172032000 }, { "epoch": 0.05, "learning_rate": 0.00047872617853560683, "loss": 3.1072, "theoretical_loss": 4.439169905332706, "tokens_seen": 172032000 }, { "epoch": 0.05, "learning_rate": 0.000478716148445336, "loss": 3.1176, "theoretical_loss": 4.438953395825102, "tokens_seen": 172097536 }, { "epoch": 0.05, "learning_rate": 0.00047870611835506524, "loss": 3.3633, "theoretical_loss": 4.438736991825744, "tokens_seen": 172163072 }, { "epoch": 0.05, "learning_rate": 0.00047869608826479437, "loss": 3.38, "theoretical_loss": 4.438520693243079, "tokens_seen": 172228608 }, { "epoch": 0.05, "learning_rate": 0.0004786860581745236, "loss": 3.0177, "theoretical_loss": 4.4383044999856756, "tokens_seen": 172294144 }, { "epoch": 0.05, "learning_rate": 0.00047867602808425273, "loss": 3.1664, "theoretical_loss": 4.438088411962211, "tokens_seen": 172359680 }, { "epoch": 0.05, "learning_rate": 0.00047866599799398197, "loss": 3.3626, "theoretical_loss": 4.437872429081477, "tokens_seen": 172425216 }, { "epoch": 0.05, "learning_rate": 0.00047865596790371115, "loss": 3.0495, "theoretical_loss": 4.437656551252381, "tokens_seen": 172490752 }, { "epoch": 0.05, "learning_rate": 0.00047864593781344033, "loss": 3.2428, "theoretical_loss": 4.4374407783839445, "tokens_seen": 172556288 }, { "epoch": 0.05, "learning_rate": 0.0004786359077231695, "loss": 3.2672, "theoretical_loss": 4.437225110385297, "tokens_seen": 172621824 }, { "epoch": 0.05, "learning_rate": 0.00047862587763289875, "loss": 3.1621, "theoretical_loss": 4.4370095471656885, "tokens_seen": 172687360 }, { "epoch": 0.05, "learning_rate": 0.0004786158475426279, "loss": 3.3215, "theoretical_loss": 4.436794088634477, "tokens_seen": 172752896 }, { "epoch": 0.05, "learning_rate": 0.0004786058174523571, "loss": 3.1152, "theoretical_loss": 4.4365787347011345, "tokens_seen": 172818432 }, { "epoch": 0.05, "learning_rate": 0.00047859578736208624, "loss": 3.1985, "theoretical_loss": 4.436363485275246, "tokens_seen": 172883968 }, { "epoch": 0.05, "learning_rate": 0.00047858575727181547, "loss": 3.3297, "theoretical_loss": 4.436148340266508, "tokens_seen": 172949504 }, { "epoch": 0.05, "learning_rate": 0.00047857572718154465, "loss": 3.3167, "theoretical_loss": 4.435933299584729, "tokens_seen": 173015040 }, { "epoch": 0.05, "learning_rate": 0.00047856569709127383, "loss": 3.0363, "theoretical_loss": 4.4357183631398325, "tokens_seen": 173080576 }, { "epoch": 0.05, "learning_rate": 0.000478555667001003, "loss": 3.1073, "theoretical_loss": 4.435503530841849, "tokens_seen": 173146112 }, { "epoch": 0.05, "learning_rate": 0.0004785456369107322, "loss": 3.1795, "theoretical_loss": 4.435288802600926, "tokens_seen": 173211648 }, { "epoch": 0.05, "learning_rate": 0.0004785356068204614, "loss": 3.4799, "theoretical_loss": 4.4350741783273175, "tokens_seen": 173277184 }, { "epoch": 0.05, "learning_rate": 0.0004785255767301906, "loss": 3.3446, "theoretical_loss": 4.434859657931392, "tokens_seen": 173342720 }, { "epoch": 0.05, "learning_rate": 0.00047851554663991974, "loss": 3.1328, "theoretical_loss": 4.434645241323629, "tokens_seen": 173408256 }, { "epoch": 0.05, "learning_rate": 0.000478505516549649, "loss": 2.8784, "theoretical_loss": 4.434430928414617, "tokens_seen": 173473792 }, { "epoch": 0.05, "learning_rate": 0.0004784954864593781, "loss": 3.2111, "theoretical_loss": 4.434216719115057, "tokens_seen": 173539328 }, { "epoch": 0.05, "learning_rate": 0.00047848545636910734, "loss": 3.3716, "theoretical_loss": 4.43400261333576, "tokens_seen": 173604864 }, { "epoch": 0.05, "objective/train/docs_used": 109645, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0553133487701416, "objective/train/theoretical_loss": 4.433788610987646, "objective/train/tokens_used": 194130400, "theoretical_loss": 4.433788610987646, "tokens_seen": 173670400 }, { "epoch": 0.05, "learning_rate": 0.0004784754262788365, "loss": 3.2501, "theoretical_loss": 4.433788610987646, "tokens_seen": 173670400 }, { "epoch": 0.05, "learning_rate": 0.0004784653961885657, "loss": 3.4322, "theoretical_loss": 4.433574711981749, "tokens_seen": 173735936 }, { "epoch": 0.05, "learning_rate": 0.0004784553660982949, "loss": 3.2243, "theoretical_loss": 4.433360916229209, "tokens_seen": 173801472 }, { "epoch": 0.05, "learning_rate": 0.0004784453360080241, "loss": 3.2569, "theoretical_loss": 4.433147223641278, "tokens_seen": 173867008 }, { "epoch": 0.05, "learning_rate": 0.00047843530591775324, "loss": 3.2601, "theoretical_loss": 4.432933634129318, "tokens_seen": 173932544 }, { "epoch": 0.05, "learning_rate": 0.0004784252758274825, "loss": 3.2892, "theoretical_loss": 4.4327201476047975, "tokens_seen": 173998080 }, { "epoch": 0.05, "learning_rate": 0.0004784152457372116, "loss": 3.3204, "theoretical_loss": 4.432506763979299, "tokens_seen": 174063616 }, { "epoch": 0.05, "learning_rate": 0.00047840521564694084, "loss": 3.3412, "theoretical_loss": 4.432293483164512, "tokens_seen": 174129152 }, { "epoch": 0.05, "learning_rate": 0.00047839518555667, "loss": 3.0976, "theoretical_loss": 4.432080305072233, "tokens_seen": 174194688 }, { "epoch": 0.05, "learning_rate": 0.0004783851554663992, "loss": 3.1193, "theoretical_loss": 4.43186722961437, "tokens_seen": 174260224 }, { "epoch": 0.05, "learning_rate": 0.0004783751253761284, "loss": 3.2963, "theoretical_loss": 4.431654256702938, "tokens_seen": 174325760 }, { "epoch": 0.05, "learning_rate": 0.00047836509528585756, "loss": 3.3016, "theoretical_loss": 4.431441386250063, "tokens_seen": 174391296 }, { "epoch": 0.05, "learning_rate": 0.00047835506519558675, "loss": 3.2971, "theoretical_loss": 4.4312286181679745, "tokens_seen": 174456832 }, { "epoch": 0.05, "learning_rate": 0.000478345035105316, "loss": 3.123, "theoretical_loss": 4.431015952369016, "tokens_seen": 174522368 }, { "epoch": 0.05, "learning_rate": 0.0004783350050150451, "loss": 3.4087, "theoretical_loss": 4.430803388765636, "tokens_seen": 174587904 }, { "epoch": 0.05, "learning_rate": 0.00047832497492477434, "loss": 3.5089, "theoretical_loss": 4.430590927270388, "tokens_seen": 174653440 }, { "epoch": 0.05, "learning_rate": 0.00047831494483450347, "loss": 3.1789, "theoretical_loss": 4.430378567795938, "tokens_seen": 174718976 }, { "epoch": 0.05, "learning_rate": 0.0004783049147442327, "loss": 3.3577, "theoretical_loss": 4.430166310255057, "tokens_seen": 174784512 }, { "epoch": 0.05, "learning_rate": 0.0004782948846539619, "loss": 3.044, "theoretical_loss": 4.429954154560624, "tokens_seen": 174850048 }, { "epoch": 0.05, "learning_rate": 0.00047828485456369107, "loss": 3.467, "theoretical_loss": 4.429742100625624, "tokens_seen": 174915584 }, { "epoch": 0.05, "learning_rate": 0.00047827482447342025, "loss": 3.0549, "theoretical_loss": 4.429530148363151, "tokens_seen": 174981120 }, { "epoch": 0.05, "learning_rate": 0.0004782647943831495, "loss": 3.323, "theoretical_loss": 4.429318297686402, "tokens_seen": 175046656 }, { "epoch": 0.05, "learning_rate": 0.0004782547642928786, "loss": 3.1715, "theoretical_loss": 4.429106548508685, "tokens_seen": 175112192 }, { "epoch": 0.05, "learning_rate": 0.00047824473420260785, "loss": 3.3082, "theoretical_loss": 4.428894900743411, "tokens_seen": 175177728 }, { "epoch": 0.05, "learning_rate": 0.000478234704112337, "loss": 3.0858, "theoretical_loss": 4.428683354304098, "tokens_seen": 175243264 }, { "epoch": 0.05, "objective/train/docs_used": 110166, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.895432710647583, "objective/train/theoretical_loss": 4.428471909104372, "objective/train/tokens_used": 195768800, "theoretical_loss": 4.428471909104372, "tokens_seen": 175308800 }, { "epoch": 0.05, "learning_rate": 0.0004782246740220662, "loss": 3.1331, "theoretical_loss": 4.428471909104372, "tokens_seen": 175308800 }, { "epoch": 0.05, "learning_rate": 0.00047821464393179544, "loss": 3.3663, "theoretical_loss": 4.428260565057964, "tokens_seen": 175374336 }, { "epoch": 0.05, "learning_rate": 0.00047820461384152457, "loss": 3.3574, "theoretical_loss": 4.428049322078708, "tokens_seen": 175439872 }, { "epoch": 0.05, "learning_rate": 0.0004781945837512538, "loss": 2.9226, "theoretical_loss": 4.427838180080547, "tokens_seen": 175505408 }, { "epoch": 0.05, "learning_rate": 0.00047818455366098293, "loss": 2.922, "theoretical_loss": 4.4276271389775275, "tokens_seen": 175570944 }, { "epoch": 0.05, "learning_rate": 0.00047817452357071217, "loss": 3.5267, "theoretical_loss": 4.427416198683803, "tokens_seen": 175636480 }, { "epoch": 0.05, "learning_rate": 0.00047816449348044135, "loss": 3.2307, "theoretical_loss": 4.427205359113629, "tokens_seen": 175702016 }, { "epoch": 0.05, "learning_rate": 0.00047815446339017053, "loss": 3.3649, "theoretical_loss": 4.42699462018137, "tokens_seen": 175767552 }, { "epoch": 0.05, "learning_rate": 0.0004781444332998997, "loss": 3.0675, "theoretical_loss": 4.42678398180149, "tokens_seen": 175833088 }, { "epoch": 0.05, "learning_rate": 0.00047813440320962895, "loss": 3.3193, "theoretical_loss": 4.426573443888563, "tokens_seen": 175898624 }, { "epoch": 0.05, "learning_rate": 0.0004781243731193581, "loss": 3.1539, "theoretical_loss": 4.426363006357263, "tokens_seen": 175964160 }, { "epoch": 0.05, "learning_rate": 0.0004781143430290873, "loss": 3.1723, "theoretical_loss": 4.426152669122374, "tokens_seen": 176029696 }, { "epoch": 0.05, "learning_rate": 0.00047810431293881644, "loss": 3.3604, "theoretical_loss": 4.425942432098774, "tokens_seen": 176095232 }, { "epoch": 0.05, "learning_rate": 0.00047809428284854567, "loss": 3.3204, "theoretical_loss": 4.425732295201455, "tokens_seen": 176160768 }, { "epoch": 0.05, "learning_rate": 0.00047808425275827485, "loss": 3.0399, "theoretical_loss": 4.425522258345508, "tokens_seen": 176226304 }, { "epoch": 0.05, "learning_rate": 0.00047807422266800403, "loss": 3.1259, "theoretical_loss": 4.425312321446127, "tokens_seen": 176291840 }, { "epoch": 0.05, "learning_rate": 0.0004780641925777332, "loss": 3.087, "theoretical_loss": 4.425102484418613, "tokens_seen": 176357376 }, { "epoch": 0.05, "learning_rate": 0.0004780541624874624, "loss": 3.0982, "theoretical_loss": 4.424892747178365, "tokens_seen": 176422912 }, { "epoch": 0.05, "learning_rate": 0.0004780441323971916, "loss": 3.1463, "theoretical_loss": 4.42468310964089, "tokens_seen": 176488448 }, { "epoch": 0.05, "learning_rate": 0.0004780341023069208, "loss": 3.0898, "theoretical_loss": 4.424473571721794, "tokens_seen": 176553984 }, { "epoch": 0.05, "learning_rate": 0.00047802407221664994, "loss": 2.6555, "theoretical_loss": 4.42426413333679, "tokens_seen": 176619520 }, { "epoch": 0.05, "learning_rate": 0.0004780140421263792, "loss": 3.364, "theoretical_loss": 4.424054794401689, "tokens_seen": 176685056 }, { "epoch": 0.05, "learning_rate": 0.0004780040120361083, "loss": 3.2562, "theoretical_loss": 4.423845554832406, "tokens_seen": 176750592 }, { "epoch": 0.05, "learning_rate": 0.00047799398194583754, "loss": 3.1273, "theoretical_loss": 4.42363641454496, "tokens_seen": 176816128 }, { "epoch": 0.05, "learning_rate": 0.0004779839518555667, "loss": 2.7713, "theoretical_loss": 4.423427373455471, "tokens_seen": 176881664 }, { "epoch": 0.05, "objective/train/docs_used": 110915, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4182002544403076, "objective/train/theoretical_loss": 4.42321843148016, "objective/train/tokens_used": 197407200, "theoretical_loss": 4.42321843148016, "tokens_seen": 176947200 }, { "epoch": 0.05, "learning_rate": 0.0004779739217652959, "loss": 3.4005, "theoretical_loss": 4.42321843148016, "tokens_seen": 176947200 }, { "epoch": 0.05, "learning_rate": 0.0004779638916750251, "loss": 3.0105, "theoretical_loss": 4.423009588535351, "tokens_seen": 177012736 }, { "epoch": 0.05, "learning_rate": 0.0004779538615847543, "loss": 3.3242, "theoretical_loss": 4.422800844537466, "tokens_seen": 177078272 }, { "epoch": 0.05, "learning_rate": 0.00047794383149448344, "loss": 3.23, "theoretical_loss": 4.422592199403036, "tokens_seen": 177143808 }, { "epoch": 0.05, "learning_rate": 0.0004779338014042127, "loss": 3.1387, "theoretical_loss": 4.422383653048685, "tokens_seen": 177209344 }, { "epoch": 0.05, "learning_rate": 0.0004779237713139418, "loss": 3.328, "theoretical_loss": 4.422175205391145, "tokens_seen": 177274880 }, { "epoch": 0.05, "learning_rate": 0.00047791374122367104, "loss": 3.1981, "theoretical_loss": 4.421966856347243, "tokens_seen": 177340416 }, { "epoch": 0.05, "learning_rate": 0.0004779037111334002, "loss": 3.1989, "theoretical_loss": 4.421758605833912, "tokens_seen": 177405952 }, { "epoch": 0.05, "learning_rate": 0.0004778936810431294, "loss": 3.0267, "theoretical_loss": 4.421550453768181, "tokens_seen": 177471488 }, { "epoch": 0.05, "learning_rate": 0.0004778836509528586, "loss": 3.4539, "theoretical_loss": 4.421342400067183, "tokens_seen": 177537024 }, { "epoch": 0.05, "learning_rate": 0.00047787362086258776, "loss": 3.2626, "theoretical_loss": 4.42113444464815, "tokens_seen": 177602560 }, { "epoch": 0.05, "learning_rate": 0.00047786359077231695, "loss": 3.4596, "theoretical_loss": 4.420926587428411, "tokens_seen": 177668096 }, { "epoch": 0.05, "learning_rate": 0.0004778535606820462, "loss": 3.1495, "theoretical_loss": 4.420718828325403, "tokens_seen": 177733632 }, { "epoch": 0.05, "learning_rate": 0.0004778435305917753, "loss": 3.2311, "theoretical_loss": 4.420511167256656, "tokens_seen": 177799168 }, { "epoch": 0.05, "learning_rate": 0.00047783350050150454, "loss": 3.2253, "theoretical_loss": 4.4203036041398, "tokens_seen": 177864704 }, { "epoch": 0.05, "learning_rate": 0.00047782347041123367, "loss": 3.1878, "theoretical_loss": 4.420096138892568, "tokens_seen": 177930240 }, { "epoch": 0.05, "learning_rate": 0.0004778134403209629, "loss": 3.0134, "theoretical_loss": 4.419888771432789, "tokens_seen": 177995776 }, { "epoch": 0.05, "learning_rate": 0.0004778034102306921, "loss": 2.9021, "theoretical_loss": 4.419681501678395, "tokens_seen": 178061312 }, { "epoch": 0.05, "learning_rate": 0.00047779338014042127, "loss": 3.3523, "theoretical_loss": 4.419474329547413, "tokens_seen": 178126848 }, { "epoch": 0.05, "learning_rate": 0.00047778335005015045, "loss": 2.9923, "theoretical_loss": 4.419267254957971, "tokens_seen": 178192384 }, { "epoch": 0.05, "learning_rate": 0.0004777733199598797, "loss": 3.1929, "theoretical_loss": 4.419060277828295, "tokens_seen": 178257920 }, { "epoch": 0.05, "learning_rate": 0.0004777632898696088, "loss": 3.4043, "theoretical_loss": 4.41885339807671, "tokens_seen": 178323456 }, { "epoch": 0.05, "learning_rate": 0.00047775325977933805, "loss": 3.2683, "theoretical_loss": 4.4186466156216415, "tokens_seen": 178388992 }, { "epoch": 0.05, "learning_rate": 0.0004777432296890672, "loss": 3.3439, "theoretical_loss": 4.418439930381609, "tokens_seen": 178454528 }, { "epoch": 0.05, "learning_rate": 0.0004777331995987964, "loss": 3.3543, "theoretical_loss": 4.418233342275233, "tokens_seen": 178520064 }, { "epoch": 0.05, "objective/train/docs_used": 112413, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1684422492980957, "objective/train/theoretical_loss": 4.418026851221231, "objective/train/tokens_used": 199045600, "theoretical_loss": 4.418026851221231, "tokens_seen": 178585600 }, { "epoch": 0.05, "learning_rate": 0.0004777231695085256, "loss": 3.2288, "theoretical_loss": 4.418026851221231, "tokens_seen": 178585600 }, { "epoch": 0.05, "learning_rate": 0.00047771313941825477, "loss": 3.31, "theoretical_loss": 4.4178204571384185, "tokens_seen": 178651136 }, { "epoch": 0.05, "learning_rate": 0.00047770310932798395, "loss": 3.2361, "theoretical_loss": 4.41761415994571, "tokens_seen": 178716672 }, { "epoch": 0.05, "learning_rate": 0.00047769307923771313, "loss": 2.9928, "theoretical_loss": 4.417407959562116, "tokens_seen": 178782208 }, { "epoch": 0.05, "learning_rate": 0.0004776830491474423, "loss": 3.21, "theoretical_loss": 4.417201855906742, "tokens_seen": 178847744 }, { "epoch": 0.05, "learning_rate": 0.00047767301905717155, "loss": 3.3201, "theoretical_loss": 4.416995848898797, "tokens_seen": 178913280 }, { "epoch": 0.05, "learning_rate": 0.0004776629889669007, "loss": 3.2827, "theoretical_loss": 4.4167899384575815, "tokens_seen": 178978816 }, { "epoch": 0.05, "learning_rate": 0.0004776529588766299, "loss": 3.3293, "theoretical_loss": 4.416584124502495, "tokens_seen": 179044352 }, { "epoch": 0.05, "learning_rate": 0.00047764292878635904, "loss": 3.3263, "theoretical_loss": 4.416378406953033, "tokens_seen": 179109888 }, { "epoch": 0.05, "learning_rate": 0.0004776328986960883, "loss": 3.3501, "theoretical_loss": 4.41617278572879, "tokens_seen": 179175424 }, { "epoch": 0.05, "learning_rate": 0.00047762286860581746, "loss": 2.9366, "theoretical_loss": 4.4159672607494524, "tokens_seen": 179240960 }, { "epoch": 0.05, "learning_rate": 0.00047761283851554664, "loss": 3.0323, "theoretical_loss": 4.415761831934808, "tokens_seen": 179306496 }, { "epoch": 0.05, "learning_rate": 0.0004776028084252758, "loss": 2.9697, "theoretical_loss": 4.415556499204737, "tokens_seen": 179372032 }, { "epoch": 0.05, "learning_rate": 0.00047759277833500505, "loss": 3.0708, "theoretical_loss": 4.415351262479216, "tokens_seen": 179437568 }, { "epoch": 0.05, "learning_rate": 0.0004775827482447342, "loss": 3.3015, "theoretical_loss": 4.415146121678321, "tokens_seen": 179503104 }, { "epoch": 0.05, "learning_rate": 0.0004775727181544634, "loss": 3.1837, "theoretical_loss": 4.414941076722219, "tokens_seen": 179568640 }, { "epoch": 0.05, "learning_rate": 0.00047756268806419254, "loss": 3.2274, "theoretical_loss": 4.4147361275311745, "tokens_seen": 179634176 }, { "epoch": 0.05, "learning_rate": 0.0004775526579739218, "loss": 3.3152, "theoretical_loss": 4.414531274025548, "tokens_seen": 179699712 }, { "epoch": 0.05, "learning_rate": 0.00047754262788365096, "loss": 3.1656, "theoretical_loss": 4.414326516125795, "tokens_seen": 179765248 }, { "epoch": 0.05, "learning_rate": 0.00047753259779338014, "loss": 3.0617, "theoretical_loss": 4.414121853752466, "tokens_seen": 179830784 }, { "epoch": 0.05, "learning_rate": 0.0004775225677031093, "loss": 3.0591, "theoretical_loss": 4.413917286826205, "tokens_seen": 179896320 }, { "epoch": 0.05, "learning_rate": 0.0004775125376128385, "loss": 3.4019, "theoretical_loss": 4.413712815267752, "tokens_seen": 179961856 }, { "epoch": 0.05, "learning_rate": 0.0004775025075225677, "loss": 3.4241, "theoretical_loss": 4.413508438997944, "tokens_seen": 180027392 }, { "epoch": 0.05, "learning_rate": 0.0004774924774322969, "loss": 2.8506, "theoretical_loss": 4.4133041579377075, "tokens_seen": 180092928 }, { "epoch": 0.05, "learning_rate": 0.00047748244734202605, "loss": 3.3765, "theoretical_loss": 4.413099972008068, "tokens_seen": 180158464 }, { "epoch": 0.05, "objective/train/docs_used": 112982, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9989802837371826, "objective/train/theoretical_loss": 4.412895881130142, "objective/train/tokens_used": 200684000, "theoretical_loss": 4.412895881130142, "tokens_seen": 180224000 }, { "epoch": 0.05, "learning_rate": 0.0004774724172517553, "loss": 3.1511, "theoretical_loss": 4.412895881130142, "tokens_seen": 180224000 }, { "epoch": 0.05, "learning_rate": 0.0004774623871614845, "loss": 3.2793, "theoretical_loss": 4.412691885225141, "tokens_seen": 180289536 }, { "epoch": 0.05, "learning_rate": 0.00047745235707121364, "loss": 3.1662, "theoretical_loss": 4.412487984214373, "tokens_seen": 180355072 }, { "epoch": 0.05, "learning_rate": 0.0004774423269809429, "loss": 2.9821, "theoretical_loss": 4.412284178019235, "tokens_seen": 180420608 }, { "epoch": 0.05, "learning_rate": 0.000477432296890672, "loss": 3.2647, "theoretical_loss": 4.412080466561221, "tokens_seen": 180486144 }, { "epoch": 0.05, "learning_rate": 0.00047742226680040124, "loss": 3.041, "theoretical_loss": 4.411876849761917, "tokens_seen": 180551680 }, { "epoch": 0.05, "learning_rate": 0.0004774122367101304, "loss": 3.229, "theoretical_loss": 4.411673327543005, "tokens_seen": 180617216 }, { "epoch": 0.05, "learning_rate": 0.0004774022066198596, "loss": 2.9898, "theoretical_loss": 4.4114698998262565, "tokens_seen": 180682752 }, { "epoch": 0.05, "learning_rate": 0.0004773921765295888, "loss": 3.4276, "theoretical_loss": 4.411266566533539, "tokens_seen": 180748288 }, { "epoch": 0.05, "learning_rate": 0.00047738214643931797, "loss": 3.1725, "theoretical_loss": 4.41106332758681, "tokens_seen": 180813824 }, { "epoch": 0.05, "learning_rate": 0.00047737211634904715, "loss": 3.1015, "theoretical_loss": 4.41086018290812, "tokens_seen": 180879360 }, { "epoch": 0.05, "learning_rate": 0.0004773620862587764, "loss": 3.1636, "theoretical_loss": 4.410657132419617, "tokens_seen": 180944896 }, { "epoch": 0.05, "learning_rate": 0.0004773520561685055, "loss": 3.2115, "theoretical_loss": 4.410454176043537, "tokens_seen": 181010432 }, { "epoch": 0.05, "learning_rate": 0.00047734202607823474, "loss": 3.1097, "theoretical_loss": 4.410251313702208, "tokens_seen": 181075968 }, { "epoch": 0.05, "learning_rate": 0.00047733199598796387, "loss": 3.322, "theoretical_loss": 4.410048545318052, "tokens_seen": 181141504 }, { "epoch": 0.05, "learning_rate": 0.0004773219658976931, "loss": 3.1962, "theoretical_loss": 4.409845870813582, "tokens_seen": 181207040 }, { "epoch": 0.05, "learning_rate": 0.0004773119358074223, "loss": 3.3378, "theoretical_loss": 4.409643290111404, "tokens_seen": 181272576 }, { "epoch": 0.05, "learning_rate": 0.00047730190571715147, "loss": 3.0122, "theoretical_loss": 4.409440803134215, "tokens_seen": 181338112 }, { "epoch": 0.05, "learning_rate": 0.00047729187562688065, "loss": 3.3294, "theoretical_loss": 4.409238409804804, "tokens_seen": 181403648 }, { "epoch": 0.05, "learning_rate": 0.0004772818455366099, "loss": 3.3616, "theoretical_loss": 4.409036110046051, "tokens_seen": 181469184 }, { "epoch": 0.06, "learning_rate": 0.000477271815446339, "loss": 2.9033, "theoretical_loss": 4.408833903780926, "tokens_seen": 181534720 }, { "epoch": 0.06, "learning_rate": 0.00047726178535606825, "loss": 3.1769, "theoretical_loss": 4.408631790932494, "tokens_seen": 181600256 }, { "epoch": 0.06, "learning_rate": 0.0004772517552657974, "loss": 3.2627, "theoretical_loss": 4.408429771423909, "tokens_seen": 181665792 }, { "epoch": 0.06, "learning_rate": 0.0004772417251755266, "loss": 3.2006, "theoretical_loss": 4.408227845178414, "tokens_seen": 181731328 }, { "epoch": 0.06, "learning_rate": 0.0004772316950852558, "loss": 3.3776, "theoretical_loss": 4.408026012119344, "tokens_seen": 181796864 }, { "epoch": 0.06, "objective/train/docs_used": 114216, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.247993230819702, "objective/train/theoretical_loss": 4.407824272170128, "objective/train/tokens_used": 202322400, "theoretical_loss": 4.407824272170128, "tokens_seen": 181862400 }, { "epoch": 0.06, "learning_rate": 0.00047722166499498497, "loss": 3.1762, "theoretical_loss": 4.407824272170128, "tokens_seen": 181862400 }, { "epoch": 0.06, "learning_rate": 0.00047721163490471415, "loss": 3.2095, "theoretical_loss": 4.407622625254279, "tokens_seen": 181927936 }, { "epoch": 0.06, "learning_rate": 0.00047720160481444333, "loss": 3.2396, "theoretical_loss": 4.407421071295406, "tokens_seen": 181993472 }, { "epoch": 0.06, "learning_rate": 0.0004771915747241725, "loss": 3.2236, "theoretical_loss": 4.407219610217206, "tokens_seen": 182059008 }, { "epoch": 0.06, "learning_rate": 0.00047718154463390175, "loss": 3.2076, "theoretical_loss": 4.407018241943467, "tokens_seen": 182124544 }, { "epoch": 0.06, "learning_rate": 0.0004771715145436309, "loss": 3.3943, "theoretical_loss": 4.406816966398064, "tokens_seen": 182190080 }, { "epoch": 0.06, "learning_rate": 0.0004771614844533601, "loss": 3.0633, "theoretical_loss": 4.406615783504965, "tokens_seen": 182255616 }, { "epoch": 0.06, "learning_rate": 0.00047715145436308924, "loss": 3.3245, "theoretical_loss": 4.4064146931882275, "tokens_seen": 182321152 }, { "epoch": 0.06, "learning_rate": 0.0004771414242728185, "loss": 3.2816, "theoretical_loss": 4.406213695371996, "tokens_seen": 182386688 }, { "epoch": 0.06, "learning_rate": 0.00047713139418254766, "loss": 3.0674, "theoretical_loss": 4.406012789980506, "tokens_seen": 182452224 }, { "epoch": 0.06, "learning_rate": 0.00047712136409227684, "loss": 3.2889, "theoretical_loss": 4.405811976938084, "tokens_seen": 182517760 }, { "epoch": 0.06, "learning_rate": 0.000477111334002006, "loss": 2.9432, "theoretical_loss": 4.405611256169143, "tokens_seen": 182583296 }, { "epoch": 0.06, "learning_rate": 0.00047710130391173525, "loss": 3.1513, "theoretical_loss": 4.405410627598185, "tokens_seen": 182648832 }, { "epoch": 0.06, "learning_rate": 0.0004770912738214644, "loss": 3.2537, "theoretical_loss": 4.405210091149802, "tokens_seen": 182714368 }, { "epoch": 0.06, "learning_rate": 0.0004770812437311936, "loss": 3.2721, "theoretical_loss": 4.405009646748674, "tokens_seen": 182779904 }, { "epoch": 0.06, "learning_rate": 0.00047707121364092274, "loss": 3.2594, "theoretical_loss": 4.404809294319572, "tokens_seen": 182845440 }, { "epoch": 0.06, "learning_rate": 0.000477061183550652, "loss": 3.0547, "theoretical_loss": 4.40460903378735, "tokens_seen": 182910976 }, { "epoch": 0.06, "learning_rate": 0.00047705115346038116, "loss": 3.238, "theoretical_loss": 4.404408865076955, "tokens_seen": 182976512 }, { "epoch": 0.06, "learning_rate": 0.00047704112337011034, "loss": 3.0375, "theoretical_loss": 4.404208788113422, "tokens_seen": 183042048 }, { "epoch": 0.06, "learning_rate": 0.0004770310932798395, "loss": 2.9986, "theoretical_loss": 4.404008802821871, "tokens_seen": 183107584 }, { "epoch": 0.06, "learning_rate": 0.0004770210631895687, "loss": 3.4418, "theoretical_loss": 4.4038089091275125, "tokens_seen": 183173120 }, { "epoch": 0.06, "learning_rate": 0.0004770110330992979, "loss": 3.1524, "theoretical_loss": 4.403609106955645, "tokens_seen": 183238656 }, { "epoch": 0.06, "learning_rate": 0.0004770010030090271, "loss": 3.4428, "theoretical_loss": 4.403409396231651, "tokens_seen": 183304192 }, { "epoch": 0.06, "learning_rate": 0.00047699097291875625, "loss": 3.097, "theoretical_loss": 4.403209776881004, "tokens_seen": 183369728 }, { "epoch": 0.06, "learning_rate": 0.0004769809428284855, "loss": 3.1748, "theoretical_loss": 4.403010248829265, "tokens_seen": 183435264 }, { "epoch": 0.06, "objective/train/docs_used": 114775, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7343997955322266, "objective/train/theoretical_loss": 4.4028108120020795, "objective/train/tokens_used": 203960800, "theoretical_loss": 4.4028108120020795, "tokens_seen": 183500800 }, { "epoch": 0.06, "learning_rate": 0.00047697091273821466, "loss": 3.364, "theoretical_loss": 4.4028108120020795, "tokens_seen": 183500800 }, { "epoch": 0.06, "learning_rate": 0.00047696088264794384, "loss": 3.5379, "theoretical_loss": 4.402611466325182, "tokens_seen": 183566336 }, { "epoch": 0.06, "learning_rate": 0.000476950852557673, "loss": 3.1437, "theoretical_loss": 4.4024122117243945, "tokens_seen": 183631872 }, { "epoch": 0.06, "learning_rate": 0.0004769408224674022, "loss": 3.0713, "theoretical_loss": 4.402213048125624, "tokens_seen": 183697408 }, { "epoch": 0.06, "learning_rate": 0.0004769307923771314, "loss": 3.1808, "theoretical_loss": 4.4020139754548655, "tokens_seen": 183762944 }, { "epoch": 0.06, "learning_rate": 0.0004769207622868606, "loss": 3.5212, "theoretical_loss": 4.401814993638199, "tokens_seen": 183828480 }, { "epoch": 0.06, "learning_rate": 0.00047691073219658975, "loss": 3.1228, "theoretical_loss": 4.4016161026017935, "tokens_seen": 183894016 }, { "epoch": 0.06, "learning_rate": 0.000476900702106319, "loss": 3.2918, "theoretical_loss": 4.401417302271902, "tokens_seen": 183959552 }, { "epoch": 0.06, "learning_rate": 0.0004768906720160481, "loss": 3.3633, "theoretical_loss": 4.401218592574865, "tokens_seen": 184025088 }, { "epoch": 0.06, "learning_rate": 0.00047688064192577735, "loss": 3.3739, "theoretical_loss": 4.401019973437108, "tokens_seen": 184090624 }, { "epoch": 0.06, "learning_rate": 0.00047687061183550653, "loss": 3.3149, "theoretical_loss": 4.400821444785143, "tokens_seen": 184156160 }, { "epoch": 0.06, "learning_rate": 0.0004768605817452357, "loss": 3.3195, "theoretical_loss": 4.400623006545567, "tokens_seen": 184221696 }, { "epoch": 0.06, "learning_rate": 0.0004768505516549649, "loss": 3.2683, "theoretical_loss": 4.400424658645065, "tokens_seen": 184287232 }, { "epoch": 0.06, "learning_rate": 0.00047684052156469407, "loss": 2.8183, "theoretical_loss": 4.400226401010404, "tokens_seen": 184352768 }, { "epoch": 0.06, "learning_rate": 0.00047683049147442325, "loss": 3.4246, "theoretical_loss": 4.40002823356844, "tokens_seen": 184418304 }, { "epoch": 0.06, "learning_rate": 0.0004768204613841525, "loss": 3.3673, "theoretical_loss": 4.39983015624611, "tokens_seen": 184483840 }, { "epoch": 0.06, "learning_rate": 0.0004768104312938816, "loss": 3.029, "theoretical_loss": 4.39963216897044, "tokens_seen": 184549376 }, { "epoch": 0.06, "learning_rate": 0.00047680040120361085, "loss": 3.2208, "theoretical_loss": 4.3994342716685395, "tokens_seen": 184614912 }, { "epoch": 0.06, "learning_rate": 0.00047679037111334003, "loss": 3.184, "theoretical_loss": 4.399236464267602, "tokens_seen": 184680448 }, { "epoch": 0.06, "learning_rate": 0.0004767803410230692, "loss": 3.2012, "theoretical_loss": 4.399038746694908, "tokens_seen": 184745984 }, { "epoch": 0.06, "learning_rate": 0.0004767703109327984, "loss": 3.4012, "theoretical_loss": 4.398841118877819, "tokens_seen": 184811520 }, { "epoch": 0.06, "learning_rate": 0.0004767602808425276, "loss": 3.0397, "theoretical_loss": 4.398643580743785, "tokens_seen": 184877056 }, { "epoch": 0.06, "learning_rate": 0.00047675025075225676, "loss": 3.3388, "theoretical_loss": 4.398446132220338, "tokens_seen": 184942592 }, { "epoch": 0.06, "learning_rate": 0.000476740220661986, "loss": 3.0237, "theoretical_loss": 4.3982487732350934, "tokens_seen": 185008128 }, { "epoch": 0.06, "learning_rate": 0.0004767301905717151, "loss": 3.1185, "theoretical_loss": 4.398051503715753, "tokens_seen": 185073664 }, { "epoch": 0.06, "objective/train/docs_used": 115939, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.676588296890259, "objective/train/theoretical_loss": 4.397854323590102, "objective/train/tokens_used": 205599200, "theoretical_loss": 4.397854323590102, "tokens_seen": 185139200 }, { "epoch": 0.06, "learning_rate": 0.00047672016048144435, "loss": 3.2948, "theoretical_loss": 4.397854323590102, "tokens_seen": 185139200 }, { "epoch": 0.06, "learning_rate": 0.00047671013039117353, "loss": 3.2664, "theoretical_loss": 4.397657232786008, "tokens_seen": 185204736 }, { "epoch": 0.06, "learning_rate": 0.0004767001003009027, "loss": 2.8907, "theoretical_loss": 4.397460231231424, "tokens_seen": 185270272 }, { "epoch": 0.06, "learning_rate": 0.00047669007021063195, "loss": 3.2102, "theoretical_loss": 4.397263318854384, "tokens_seen": 185335808 }, { "epoch": 0.06, "learning_rate": 0.0004766800401203611, "loss": 3.0437, "theoretical_loss": 4.39706649558301, "tokens_seen": 185401344 }, { "epoch": 0.06, "learning_rate": 0.0004766700100300903, "loss": 3.2762, "theoretical_loss": 4.396869761345503, "tokens_seen": 185466880 }, { "epoch": 0.06, "learning_rate": 0.00047665997993981944, "loss": 3.2785, "theoretical_loss": 4.396673116070147, "tokens_seen": 185532416 }, { "epoch": 0.06, "learning_rate": 0.0004766499498495487, "loss": 3.169, "theoretical_loss": 4.396476559685315, "tokens_seen": 185597952 }, { "epoch": 0.06, "learning_rate": 0.00047663991975927786, "loss": 2.8861, "theoretical_loss": 4.396280092119455, "tokens_seen": 185663488 }, { "epoch": 0.06, "learning_rate": 0.00047662988966900704, "loss": 3.6408, "theoretical_loss": 4.3960837133011035, "tokens_seen": 185729024 }, { "epoch": 0.06, "learning_rate": 0.0004766198595787362, "loss": 3.29, "theoretical_loss": 4.395887423158877, "tokens_seen": 185794560 }, { "epoch": 0.06, "learning_rate": 0.00047660982948846545, "loss": 3.2726, "theoretical_loss": 4.395691221621476, "tokens_seen": 185860096 }, { "epoch": 0.06, "learning_rate": 0.0004765997993981946, "loss": 3.2641, "theoretical_loss": 4.395495108617682, "tokens_seen": 185925632 }, { "epoch": 0.06, "learning_rate": 0.0004765897693079238, "loss": 3.3623, "theoretical_loss": 4.39529908407636, "tokens_seen": 185991168 }, { "epoch": 0.06, "learning_rate": 0.00047657973921765294, "loss": 3.0954, "theoretical_loss": 4.3951031479264575, "tokens_seen": 186056704 }, { "epoch": 0.06, "learning_rate": 0.0004765697091273822, "loss": 3.4236, "theoretical_loss": 4.394907300097002, "tokens_seen": 186122240 }, { "epoch": 0.06, "learning_rate": 0.00047655967903711136, "loss": 3.328, "theoretical_loss": 4.394711540517106, "tokens_seen": 186187776 }, { "epoch": 0.06, "learning_rate": 0.00047654964894684054, "loss": 3.3462, "theoretical_loss": 4.39451586911596, "tokens_seen": 186253312 }, { "epoch": 0.06, "learning_rate": 0.0004765396188565697, "loss": 3.3489, "theoretical_loss": 4.39432028582284, "tokens_seen": 186318848 }, { "epoch": 0.06, "learning_rate": 0.0004765295887662989, "loss": 3.2442, "theoretical_loss": 4.394124790567101, "tokens_seen": 186384384 }, { "epoch": 0.06, "learning_rate": 0.0004765195586760281, "loss": 3.6337, "theoretical_loss": 4.3939293832781825, "tokens_seen": 186449920 }, { "epoch": 0.06, "learning_rate": 0.0004765095285857573, "loss": 3.2979, "theoretical_loss": 4.393734063885599, "tokens_seen": 186515456 }, { "epoch": 0.06, "learning_rate": 0.00047649949849548645, "loss": 3.1169, "theoretical_loss": 4.3935388323189555, "tokens_seen": 186580992 }, { "epoch": 0.06, "learning_rate": 0.0004764894684052157, "loss": 3.2682, "theoretical_loss": 4.39334368850793, "tokens_seen": 186646528 }, { "epoch": 0.06, "learning_rate": 0.00047647943831494486, "loss": 3.1131, "theoretical_loss": 4.3931486323822835, "tokens_seen": 186712064 }, { "epoch": 0.06, "objective/train/docs_used": 116562, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.736578941345215, "objective/train/theoretical_loss": 4.392953663871862, "objective/train/tokens_used": 207237600, "theoretical_loss": 4.392953663871862, "tokens_seen": 186777600 }, { "epoch": 0.06, "learning_rate": 0.00047646940822467404, "loss": 3.3584, "theoretical_loss": 4.392953663871862, "tokens_seen": 186777600 }, { "epoch": 0.06, "learning_rate": 0.0004764593781344032, "loss": 3.162, "theoretical_loss": 4.392758782906586, "tokens_seen": 186843136 }, { "epoch": 0.06, "learning_rate": 0.0004764493480441324, "loss": 3.062, "theoretical_loss": 4.392563989416462, "tokens_seen": 186908672 }, { "epoch": 0.06, "learning_rate": 0.0004764393179538616, "loss": 3.478, "theoretical_loss": 4.392369283331574, "tokens_seen": 186974208 }, { "epoch": 0.06, "learning_rate": 0.0004764292878635908, "loss": 3.2393, "theoretical_loss": 4.392174664582085, "tokens_seen": 187039744 }, { "epoch": 0.06, "learning_rate": 0.00047641925777331995, "loss": 3.1537, "theoretical_loss": 4.391980133098244, "tokens_seen": 187105280 }, { "epoch": 0.06, "learning_rate": 0.0004764092276830492, "loss": 3.2428, "theoretical_loss": 4.391785688810373, "tokens_seen": 187170816 }, { "epoch": 0.06, "learning_rate": 0.0004763991975927783, "loss": 3.2938, "theoretical_loss": 4.391591331648879, "tokens_seen": 187236352 }, { "epoch": 0.06, "learning_rate": 0.00047638916750250755, "loss": 3.4311, "theoretical_loss": 4.391397061544247, "tokens_seen": 187301888 }, { "epoch": 0.06, "learning_rate": 0.00047637913741223673, "loss": 3.0581, "theoretical_loss": 4.391202878427042, "tokens_seen": 187367424 }, { "epoch": 0.06, "learning_rate": 0.0004763691073219659, "loss": 3.5038, "theoretical_loss": 4.3910087822279085, "tokens_seen": 187432960 }, { "epoch": 0.06, "learning_rate": 0.0004763590772316951, "loss": 3.2592, "theoretical_loss": 4.390814772877571, "tokens_seen": 187498496 }, { "epoch": 0.06, "learning_rate": 0.00047634904714142427, "loss": 3.0196, "theoretical_loss": 4.390620850306832, "tokens_seen": 187564032 }, { "epoch": 0.06, "learning_rate": 0.00047633901705115345, "loss": 3.2439, "theoretical_loss": 4.390427014446575, "tokens_seen": 187629568 }, { "epoch": 0.06, "learning_rate": 0.0004763289869608827, "loss": 3.3844, "theoretical_loss": 4.390233265227764, "tokens_seen": 187695104 }, { "epoch": 0.06, "learning_rate": 0.0004763189568706118, "loss": 3.1343, "theoretical_loss": 4.390039602581437, "tokens_seen": 187760640 }, { "epoch": 0.06, "learning_rate": 0.00047630892678034105, "loss": 2.9602, "theoretical_loss": 4.389846026438715, "tokens_seen": 187826176 }, { "epoch": 0.06, "learning_rate": 0.00047629889669007023, "loss": 3.2549, "theoretical_loss": 4.3896525367307975, "tokens_seen": 187891712 }, { "epoch": 0.06, "learning_rate": 0.0004762888665997994, "loss": 2.9531, "theoretical_loss": 4.389459133388962, "tokens_seen": 187957248 }, { "epoch": 0.06, "learning_rate": 0.0004762788365095286, "loss": 3.1851, "theoretical_loss": 4.3892658163445635, "tokens_seen": 188022784 }, { "epoch": 0.06, "learning_rate": 0.0004762688064192578, "loss": 3.2693, "theoretical_loss": 4.389072585529037, "tokens_seen": 188088320 }, { "epoch": 0.06, "learning_rate": 0.00047625877632898696, "loss": 3.1727, "theoretical_loss": 4.388879440873897, "tokens_seen": 188153856 }, { "epoch": 0.06, "learning_rate": 0.0004762487462387162, "loss": 3.3616, "theoretical_loss": 4.388686382310732, "tokens_seen": 188219392 }, { "epoch": 0.06, "learning_rate": 0.0004762387161484453, "loss": 3.176, "theoretical_loss": 4.388493409771213, "tokens_seen": 188284928 }, { "epoch": 0.06, "learning_rate": 0.00047622868605817455, "loss": 3.135, "theoretical_loss": 4.388300523187087, "tokens_seen": 188350464 }, { "epoch": 0.06, "objective/train/docs_used": 117858, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.447967290878296, "objective/train/theoretical_loss": 4.3881077224901786, "objective/train/tokens_used": 208876000, "theoretical_loss": 4.3881077224901786, "tokens_seen": 188416000 }, { "epoch": 0.06, "learning_rate": 0.0004762186559679037, "loss": 3.4454, "theoretical_loss": 4.3881077224901786, "tokens_seen": 188416000 }, { "epoch": 0.06, "learning_rate": 0.0004762086258776329, "loss": 3.1037, "theoretical_loss": 4.38791500761239, "tokens_seen": 188481536 }, { "epoch": 0.06, "learning_rate": 0.0004761985957873621, "loss": 3.3634, "theoretical_loss": 4.387722378485703, "tokens_seen": 188547072 }, { "epoch": 0.06, "learning_rate": 0.0004761885656970913, "loss": 3.1468, "theoretical_loss": 4.3875298350421765, "tokens_seen": 188612608 }, { "epoch": 0.06, "learning_rate": 0.00047617853560682046, "loss": 3.3193, "theoretical_loss": 4.387337377213943, "tokens_seen": 188678144 }, { "epoch": 0.06, "learning_rate": 0.00047616850551654964, "loss": 3.2997, "theoretical_loss": 4.387145004933218, "tokens_seen": 188743680 }, { "epoch": 0.06, "learning_rate": 0.0004761584754262788, "loss": 3.0325, "theoretical_loss": 4.38695271813229, "tokens_seen": 188809216 }, { "epoch": 0.06, "learning_rate": 0.00047614844533600806, "loss": 3.0921, "theoretical_loss": 4.386760516743526, "tokens_seen": 188874752 }, { "epoch": 0.06, "learning_rate": 0.0004761384152457372, "loss": 3.3926, "theoretical_loss": 4.38656840069937, "tokens_seen": 188940288 }, { "epoch": 0.06, "learning_rate": 0.0004761283851554664, "loss": 3.1398, "theoretical_loss": 4.386376369932344, "tokens_seen": 189005824 }, { "epoch": 0.06, "learning_rate": 0.0004761183550651956, "loss": 3.0158, "theoretical_loss": 4.386184424375044, "tokens_seen": 189071360 }, { "epoch": 0.06, "learning_rate": 0.0004761083249749248, "loss": 3.4574, "theoretical_loss": 4.385992563960145, "tokens_seen": 189136896 }, { "epoch": 0.06, "learning_rate": 0.00047609829488465396, "loss": 3.1658, "theoretical_loss": 4.385800788620397, "tokens_seen": 189202432 }, { "epoch": 0.06, "learning_rate": 0.00047608826479438314, "loss": 3.1425, "theoretical_loss": 4.385609098288628, "tokens_seen": 189267968 }, { "epoch": 0.06, "learning_rate": 0.0004760782347041123, "loss": 3.239, "theoretical_loss": 4.385417492897741, "tokens_seen": 189333504 }, { "epoch": 0.06, "learning_rate": 0.00047606820461384156, "loss": 3.3172, "theoretical_loss": 4.385225972380715, "tokens_seen": 189399040 }, { "epoch": 0.06, "learning_rate": 0.0004760581745235707, "loss": 3.2791, "theoretical_loss": 4.385034536670606, "tokens_seen": 189464576 }, { "epoch": 0.06, "learning_rate": 0.0004760481444332999, "loss": 3.33, "theoretical_loss": 4.384843185700544, "tokens_seen": 189530112 }, { "epoch": 0.06, "learning_rate": 0.00047603811434302905, "loss": 3.3761, "theoretical_loss": 4.384651919403739, "tokens_seen": 189595648 }, { "epoch": 0.06, "learning_rate": 0.0004760280842527583, "loss": 3.2921, "theoretical_loss": 4.384460737713471, "tokens_seen": 189661184 }, { "epoch": 0.06, "learning_rate": 0.00047601805416248746, "loss": 3.4782, "theoretical_loss": 4.384269640563101, "tokens_seen": 189726720 }, { "epoch": 0.06, "learning_rate": 0.00047600802407221665, "loss": 3.1232, "theoretical_loss": 4.384078627886062, "tokens_seen": 189792256 }, { "epoch": 0.06, "learning_rate": 0.0004759979939819458, "loss": 3.2606, "theoretical_loss": 4.383887699615863, "tokens_seen": 189857792 }, { "epoch": 0.06, "learning_rate": 0.00047598796389167506, "loss": 3.0839, "theoretical_loss": 4.38369685568609, "tokens_seen": 189923328 }, { "epoch": 0.06, "learning_rate": 0.0004759779338014042, "loss": 2.9678, "theoretical_loss": 4.383506096030401, "tokens_seen": 189988864 }, { "epoch": 0.06, "objective/train/docs_used": 118609, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9445102214813232, "objective/train/theoretical_loss": 4.383315420582533, "objective/train/tokens_used": 210514400, "theoretical_loss": 4.383315420582533, "tokens_seen": 190054400 }, { "epoch": 0.06, "learning_rate": 0.0004759679037111334, "loss": 3.2935, "theoretical_loss": 4.383315420582533, "tokens_seen": 190054400 }, { "epoch": 0.06, "learning_rate": 0.0004759578736208626, "loss": 3.2186, "theoretical_loss": 4.383124829276294, "tokens_seen": 190119936 }, { "epoch": 0.06, "learning_rate": 0.0004759478435305918, "loss": 3.2593, "theoretical_loss": 4.38293432204557, "tokens_seen": 190185472 }, { "epoch": 0.06, "learning_rate": 0.000475937813440321, "loss": 3.1491, "theoretical_loss": 4.382743898824321, "tokens_seen": 190251008 }, { "epoch": 0.06, "learning_rate": 0.00047592778335005015, "loss": 3.1876, "theoretical_loss": 4.3825535595465785, "tokens_seen": 190316544 }, { "epoch": 0.06, "learning_rate": 0.0004759177532597794, "loss": 3.3843, "theoretical_loss": 4.382363304146453, "tokens_seen": 190382080 }, { "epoch": 0.06, "learning_rate": 0.0004759077231695085, "loss": 3.2096, "theoretical_loss": 4.382173132558126, "tokens_seen": 190447616 }, { "epoch": 0.06, "learning_rate": 0.00047589769307923775, "loss": 3.0433, "theoretical_loss": 4.381983044715856, "tokens_seen": 190513152 }, { "epoch": 0.06, "learning_rate": 0.00047588766298896693, "loss": 3.2308, "theoretical_loss": 4.381793040553973, "tokens_seen": 190578688 }, { "epoch": 0.06, "learning_rate": 0.0004758776328986961, "loss": 3.478, "theoretical_loss": 4.381603120006883, "tokens_seen": 190644224 }, { "epoch": 0.06, "learning_rate": 0.0004758676028084253, "loss": 3.4116, "theoretical_loss": 4.381413283009065, "tokens_seen": 190709760 }, { "epoch": 0.06, "learning_rate": 0.00047585757271815447, "loss": 3.0949, "theoretical_loss": 4.381223529495073, "tokens_seen": 190775296 }, { "epoch": 0.06, "learning_rate": 0.00047584754262788365, "loss": 3.1533, "theoretical_loss": 4.381033859399532, "tokens_seen": 190840832 }, { "epoch": 0.06, "learning_rate": 0.0004758375125376129, "loss": 3.2304, "theoretical_loss": 4.380844272657145, "tokens_seen": 190906368 }, { "epoch": 0.06, "learning_rate": 0.000475827482447342, "loss": 3.3131, "theoretical_loss": 4.380654769202683, "tokens_seen": 190971904 }, { "epoch": 0.06, "learning_rate": 0.00047581745235707125, "loss": 2.9842, "theoretical_loss": 4.380465348970995, "tokens_seen": 191037440 }, { "epoch": 0.06, "learning_rate": 0.00047580742226680043, "loss": 3.1313, "theoretical_loss": 4.380276011897003, "tokens_seen": 191102976 }, { "epoch": 0.06, "learning_rate": 0.0004757973921765296, "loss": 3.2165, "theoretical_loss": 4.380086757915698, "tokens_seen": 191168512 }, { "epoch": 0.06, "learning_rate": 0.0004757873620862588, "loss": 3.2402, "theoretical_loss": 4.379897586962148, "tokens_seen": 191234048 }, { "epoch": 0.06, "learning_rate": 0.000475777331995988, "loss": 3.1014, "theoretical_loss": 4.379708498971494, "tokens_seen": 191299584 }, { "epoch": 0.06, "learning_rate": 0.00047576730190571716, "loss": 3.2106, "theoretical_loss": 4.379519493878948, "tokens_seen": 191365120 }, { "epoch": 0.06, "learning_rate": 0.0004757572718154464, "loss": 3.2548, "theoretical_loss": 4.379330571619795, "tokens_seen": 191430656 }, { "epoch": 0.06, "learning_rate": 0.0004757472417251755, "loss": 3.3656, "theoretical_loss": 4.379141732129394, "tokens_seen": 191496192 }, { "epoch": 0.06, "learning_rate": 0.00047573721163490475, "loss": 3.4178, "theoretical_loss": 4.378952975343175, "tokens_seen": 191561728 }, { "epoch": 0.06, "learning_rate": 0.0004757271815446339, "loss": 3.1327, "theoretical_loss": 4.378764301196642, "tokens_seen": 191627264 }, { "epoch": 0.06, "objective/train/docs_used": 119339, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7788050174713135, "objective/train/theoretical_loss": 4.37857570962537, "objective/train/tokens_used": 212152800, "theoretical_loss": 4.37857570962537, "tokens_seen": 191692800 }, { "epoch": 0.06, "learning_rate": 0.0004757171514543631, "loss": 3.0324, "theoretical_loss": 4.37857570962537, "tokens_seen": 191692800 }, { "epoch": 0.06, "learning_rate": 0.0004757071213640923, "loss": 3.1974, "theoretical_loss": 4.378387200565006, "tokens_seen": 191758336 }, { "epoch": 0.06, "learning_rate": 0.0004756970912738215, "loss": 3.4038, "theoretical_loss": 4.378198773951272, "tokens_seen": 191823872 }, { "epoch": 0.06, "learning_rate": 0.00047568706118355066, "loss": 3.0394, "theoretical_loss": 4.378010429719957, "tokens_seen": 191889408 }, { "epoch": 0.06, "learning_rate": 0.00047567703109327984, "loss": 2.9882, "theoretical_loss": 4.377822167806928, "tokens_seen": 191954944 }, { "epoch": 0.06, "learning_rate": 0.000475667001003009, "loss": 3.0199, "theoretical_loss": 4.377633988148117, "tokens_seen": 192020480 }, { "epoch": 0.06, "learning_rate": 0.00047565697091273826, "loss": 3.2158, "theoretical_loss": 4.377445890679534, "tokens_seen": 192086016 }, { "epoch": 0.06, "learning_rate": 0.0004756469408224674, "loss": 3.0202, "theoretical_loss": 4.377257875337257, "tokens_seen": 192151552 }, { "epoch": 0.06, "learning_rate": 0.0004756369107321966, "loss": 3.1161, "theoretical_loss": 4.377069942057436, "tokens_seen": 192217088 }, { "epoch": 0.06, "learning_rate": 0.0004756268806419258, "loss": 3.2261, "theoretical_loss": 4.376882090776293, "tokens_seen": 192282624 }, { "epoch": 0.06, "learning_rate": 0.000475616850551655, "loss": 3.3294, "theoretical_loss": 4.376694321430121, "tokens_seen": 192348160 }, { "epoch": 0.06, "learning_rate": 0.00047560682046138416, "loss": 3.2889, "theoretical_loss": 4.376506633955286, "tokens_seen": 192413696 }, { "epoch": 0.06, "learning_rate": 0.00047559679037111334, "loss": 3.3172, "theoretical_loss": 4.376319028288219, "tokens_seen": 192479232 }, { "epoch": 0.06, "learning_rate": 0.0004755867602808425, "loss": 3.426, "theoretical_loss": 4.37613150436543, "tokens_seen": 192544768 }, { "epoch": 0.06, "learning_rate": 0.00047557673019057176, "loss": 3.0088, "theoretical_loss": 4.375944062123496, "tokens_seen": 192610304 }, { "epoch": 0.06, "learning_rate": 0.0004755667001003009, "loss": 3.4675, "theoretical_loss": 4.375756701499063, "tokens_seen": 192675840 }, { "epoch": 0.06, "learning_rate": 0.0004755566700100301, "loss": 3.2118, "theoretical_loss": 4.3755694224288515, "tokens_seen": 192741376 }, { "epoch": 0.06, "learning_rate": 0.00047554663991975925, "loss": 3.1241, "theoretical_loss": 4.375382224849648, "tokens_seen": 192806912 }, { "epoch": 0.06, "learning_rate": 0.0004755366098294885, "loss": 3.2397, "theoretical_loss": 4.375195108698316, "tokens_seen": 192872448 }, { "epoch": 0.06, "learning_rate": 0.00047552657973921766, "loss": 3.6427, "theoretical_loss": 4.375008073911781, "tokens_seen": 192937984 }, { "epoch": 0.06, "learning_rate": 0.00047551654964894685, "loss": 3.1663, "theoretical_loss": 4.374821120427047, "tokens_seen": 193003520 }, { "epoch": 0.06, "learning_rate": 0.00047550651955867603, "loss": 3.0961, "theoretical_loss": 4.374634248181182, "tokens_seen": 193069056 }, { "epoch": 0.06, "learning_rate": 0.00047549648946840526, "loss": 3.5514, "theoretical_loss": 4.3744474571113265, "tokens_seen": 193134592 }, { "epoch": 0.06, "learning_rate": 0.0004754864593781344, "loss": 3.0639, "theoretical_loss": 4.374260747154692, "tokens_seen": 193200128 }, { "epoch": 0.06, "learning_rate": 0.0004754764292878636, "loss": 3.178, "theoretical_loss": 4.374074118248559, "tokens_seen": 193265664 }, { "epoch": 0.06, "objective/train/docs_used": 120055, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.8682849407196045, "objective/train/theoretical_loss": 4.373887570330275, "objective/train/tokens_used": 213791200, "theoretical_loss": 4.373887570330275, "tokens_seen": 193331200 }, { "epoch": 0.06, "learning_rate": 0.00047546639919759275, "loss": 3.3739, "theoretical_loss": 4.373887570330275, "tokens_seen": 193331200 }, { "epoch": 0.06, "learning_rate": 0.000475456369107322, "loss": 3.3962, "theoretical_loss": 4.373701103337263, "tokens_seen": 193396736 }, { "epoch": 0.06, "learning_rate": 0.00047544633901705117, "loss": 3.1677, "theoretical_loss": 4.373514717207009, "tokens_seen": 193462272 }, { "epoch": 0.06, "learning_rate": 0.00047543630892678035, "loss": 3.2994, "theoretical_loss": 4.373328411877073, "tokens_seen": 193527808 }, { "epoch": 0.06, "learning_rate": 0.00047542627883650953, "loss": 3.3462, "theoretical_loss": 4.373142187285083, "tokens_seen": 193593344 }, { "epoch": 0.06, "learning_rate": 0.0004754162487462387, "loss": 3.3214, "theoretical_loss": 4.372956043368736, "tokens_seen": 193658880 }, { "epoch": 0.06, "learning_rate": 0.0004754062186559679, "loss": 3.333, "theoretical_loss": 4.372769980065797, "tokens_seen": 193724416 }, { "epoch": 0.06, "learning_rate": 0.00047539618856569713, "loss": 3.3139, "theoretical_loss": 4.372583997314104, "tokens_seen": 193789952 }, { "epoch": 0.06, "learning_rate": 0.00047538615847542625, "loss": 3.2031, "theoretical_loss": 4.372398095051559, "tokens_seen": 193855488 }, { "epoch": 0.06, "learning_rate": 0.0004753761283851555, "loss": 3.3055, "theoretical_loss": 4.372212273216136, "tokens_seen": 193921024 }, { "epoch": 0.06, "learning_rate": 0.0004753660982948846, "loss": 3.3132, "theoretical_loss": 4.372026531745877, "tokens_seen": 193986560 }, { "epoch": 0.06, "learning_rate": 0.00047535606820461385, "loss": 3.2799, "theoretical_loss": 4.371840870578891, "tokens_seen": 194052096 }, { "epoch": 0.06, "learning_rate": 0.00047534603811434303, "loss": 3.0888, "theoretical_loss": 4.37165528965336, "tokens_seen": 194117632 }, { "epoch": 0.06, "learning_rate": 0.0004753360080240722, "loss": 3.441, "theoretical_loss": 4.371469788907529, "tokens_seen": 194183168 }, { "epoch": 0.06, "learning_rate": 0.0004753259779338014, "loss": 3.1359, "theoretical_loss": 4.371284368279714, "tokens_seen": 194248704 }, { "epoch": 0.06, "learning_rate": 0.00047531594784353063, "loss": 3.282, "theoretical_loss": 4.3710990277083, "tokens_seen": 194314240 }, { "epoch": 0.06, "learning_rate": 0.00047530591775325976, "loss": 3.1125, "theoretical_loss": 4.3709137671317375, "tokens_seen": 194379776 }, { "epoch": 0.06, "learning_rate": 0.000475295887662989, "loss": 3.1613, "theoretical_loss": 4.37072858648855, "tokens_seen": 194445312 }, { "epoch": 0.06, "learning_rate": 0.0004752858575727181, "loss": 3.267, "theoretical_loss": 4.370543485717322, "tokens_seen": 194510848 }, { "epoch": 0.06, "learning_rate": 0.00047527582748244736, "loss": 3.2022, "theoretical_loss": 4.370358464756713, "tokens_seen": 194576384 }, { "epoch": 0.06, "learning_rate": 0.00047526579739217654, "loss": 3.2746, "theoretical_loss": 4.370173523545443, "tokens_seen": 194641920 }, { "epoch": 0.06, "learning_rate": 0.0004752557673019057, "loss": 3.3689, "theoretical_loss": 4.3699886620223065, "tokens_seen": 194707456 }, { "epoch": 0.06, "learning_rate": 0.0004752457372116349, "loss": 3.4246, "theoretical_loss": 4.369803880126162, "tokens_seen": 194772992 }, { "epoch": 0.06, "learning_rate": 0.0004752357071213641, "loss": 3.0546, "theoretical_loss": 4.3696191777959354, "tokens_seen": 194838528 }, { "epoch": 0.06, "learning_rate": 0.00047522567703109326, "loss": 3.254, "theoretical_loss": 4.369434554970621, "tokens_seen": 194904064 }, { "epoch": 0.06, "objective/train/docs_used": 121265, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3207590579986572, "objective/train/theoretical_loss": 4.369250011589279, "objective/train/tokens_used": 215429600, "theoretical_loss": 4.369250011589279, "tokens_seen": 194969600 }, { "epoch": 0.06, "learning_rate": 0.0004752156469408225, "loss": 3.1958, "theoretical_loss": 4.369250011589279, "tokens_seen": 194969600 }, { "epoch": 0.06, "learning_rate": 0.0004752056168505517, "loss": 3.1648, "theoretical_loss": 4.369065547591038, "tokens_seen": 195035136 }, { "epoch": 0.06, "learning_rate": 0.00047519558676028086, "loss": 3.2667, "theoretical_loss": 4.368881162915095, "tokens_seen": 195100672 }, { "epoch": 0.06, "learning_rate": 0.00047518555667001004, "loss": 3.3682, "theoretical_loss": 4.36869685750071, "tokens_seen": 195166208 }, { "epoch": 0.06, "learning_rate": 0.0004751755265797392, "loss": 3.4655, "theoretical_loss": 4.3685126312872145, "tokens_seen": 195231744 }, { "epoch": 0.06, "learning_rate": 0.00047516549648946846, "loss": 3.0632, "theoretical_loss": 4.368328484214002, "tokens_seen": 195297280 }, { "epoch": 0.06, "learning_rate": 0.0004751554663991976, "loss": 3.2254, "theoretical_loss": 4.368144416220538, "tokens_seen": 195362816 }, { "epoch": 0.06, "learning_rate": 0.0004751454363089268, "loss": 3.26, "theoretical_loss": 4.3679604272463495, "tokens_seen": 195428352 }, { "epoch": 0.06, "learning_rate": 0.000475135406218656, "loss": 3.19, "theoretical_loss": 4.367776517231033, "tokens_seen": 195493888 }, { "epoch": 0.06, "learning_rate": 0.0004751253761283852, "loss": 3.282, "theoretical_loss": 4.367592686114252, "tokens_seen": 195559424 }, { "epoch": 0.06, "learning_rate": 0.00047511534603811436, "loss": 3.1387, "theoretical_loss": 4.367408933835733, "tokens_seen": 195624960 }, { "epoch": 0.06, "learning_rate": 0.00047510531594784354, "loss": 3.3084, "theoretical_loss": 4.367225260335272, "tokens_seen": 195690496 }, { "epoch": 0.06, "learning_rate": 0.0004750952858575727, "loss": 3.409, "theoretical_loss": 4.36704166555273, "tokens_seen": 195756032 }, { "epoch": 0.06, "learning_rate": 0.00047508525576730196, "loss": 3.3344, "theoretical_loss": 4.366858149428032, "tokens_seen": 195821568 }, { "epoch": 0.06, "learning_rate": 0.0004750752256770311, "loss": 3.0945, "theoretical_loss": 4.366674711901173, "tokens_seen": 195887104 }, { "epoch": 0.06, "learning_rate": 0.0004750651955867603, "loss": 3.1295, "theoretical_loss": 4.366491352912211, "tokens_seen": 195952640 }, { "epoch": 0.06, "learning_rate": 0.00047505516549648945, "loss": 3.1857, "theoretical_loss": 4.366308072401271, "tokens_seen": 196018176 }, { "epoch": 0.06, "learning_rate": 0.0004750451354062187, "loss": 3.2574, "theoretical_loss": 4.366124870308541, "tokens_seen": 196083712 }, { "epoch": 0.06, "learning_rate": 0.00047503510531594787, "loss": 3.1826, "theoretical_loss": 4.365941746574278, "tokens_seen": 196149248 }, { "epoch": 0.06, "learning_rate": 0.00047502507522567705, "loss": 3.1245, "theoretical_loss": 4.3657587011388035, "tokens_seen": 196214784 }, { "epoch": 0.06, "learning_rate": 0.00047501504513540623, "loss": 3.1448, "theoretical_loss": 4.365575733942503, "tokens_seen": 196280320 }, { "epoch": 0.06, "learning_rate": 0.00047500501504513546, "loss": 3.1648, "theoretical_loss": 4.365392844925829, "tokens_seen": 196345856 }, { "epoch": 0.06, "learning_rate": 0.0004749949849548646, "loss": 3.2024, "theoretical_loss": 4.365210034029298, "tokens_seen": 196411392 }, { "epoch": 0.06, "learning_rate": 0.0004749849548645938, "loss": 3.0918, "theoretical_loss": 4.365027301193491, "tokens_seen": 196476928 }, { "epoch": 0.06, "learning_rate": 0.00047497492477432295, "loss": 2.8404, "theoretical_loss": 4.364844646359056, "tokens_seen": 196542464 }, { "epoch": 0.06, "objective/train/docs_used": 121711, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7159526348114014, "objective/train/theoretical_loss": 4.364662069466704, "objective/train/tokens_used": 217068000, "theoretical_loss": 4.364662069466704, "tokens_seen": 196608000 }, { "epoch": 0.06, "learning_rate": 0.0004749648946840522, "loss": 3.3992, "theoretical_loss": 4.364662069466704, "tokens_seen": 196608000 }, { "epoch": 0.06, "learning_rate": 0.00047495486459378137, "loss": 3.3754, "theoretical_loss": 4.364479570457213, "tokens_seen": 196673536 }, { "epoch": 0.06, "learning_rate": 0.00047494483450351055, "loss": 3.1078, "theoretical_loss": 4.364297149271423, "tokens_seen": 196739072 }, { "epoch": 0.06, "learning_rate": 0.00047493480441323973, "loss": 3.2097, "theoretical_loss": 4.3641148058502415, "tokens_seen": 196804608 }, { "epoch": 0.06, "learning_rate": 0.0004749247743229689, "loss": 3.0253, "theoretical_loss": 4.363932540134638, "tokens_seen": 196870144 }, { "epoch": 0.06, "learning_rate": 0.0004749147442326981, "loss": 3.4151, "theoretical_loss": 4.363750352065647, "tokens_seen": 196935680 }, { "epoch": 0.06, "learning_rate": 0.00047490471414242733, "loss": 3.1998, "theoretical_loss": 4.363568241584368, "tokens_seen": 197001216 }, { "epoch": 0.06, "learning_rate": 0.00047489468405215646, "loss": 3.2949, "theoretical_loss": 4.363386208631966, "tokens_seen": 197066752 }, { "epoch": 0.06, "learning_rate": 0.0004748846539618857, "loss": 3.153, "theoretical_loss": 4.363204253149667, "tokens_seen": 197132288 }, { "epoch": 0.06, "learning_rate": 0.0004748746238716148, "loss": 3.3637, "theoretical_loss": 4.3630223750787644, "tokens_seen": 197197824 }, { "epoch": 0.06, "learning_rate": 0.00047486459378134405, "loss": 3.0955, "theoretical_loss": 4.362840574360612, "tokens_seen": 197263360 }, { "epoch": 0.06, "learning_rate": 0.00047485456369107323, "loss": 3.2066, "theoretical_loss": 4.362658850936631, "tokens_seen": 197328896 }, { "epoch": 0.06, "learning_rate": 0.0004748445336008024, "loss": 3.065, "theoretical_loss": 4.362477204748305, "tokens_seen": 197394432 }, { "epoch": 0.06, "learning_rate": 0.0004748345035105316, "loss": 3.1457, "theoretical_loss": 4.362295635737179, "tokens_seen": 197459968 }, { "epoch": 0.06, "learning_rate": 0.00047482447342026083, "loss": 3.2047, "theoretical_loss": 4.362114143844867, "tokens_seen": 197525504 }, { "epoch": 0.06, "learning_rate": 0.00047481444332998996, "loss": 3.2647, "theoretical_loss": 4.3619327290130405, "tokens_seen": 197591040 }, { "epoch": 0.06, "learning_rate": 0.0004748044132397192, "loss": 3.3005, "theoretical_loss": 4.3617513911834385, "tokens_seen": 197656576 }, { "epoch": 0.06, "learning_rate": 0.0004747943831494483, "loss": 3.1384, "theoretical_loss": 4.361570130297863, "tokens_seen": 197722112 }, { "epoch": 0.06, "learning_rate": 0.00047478435305917756, "loss": 3.3974, "theoretical_loss": 4.3613889462981765, "tokens_seen": 197787648 }, { "epoch": 0.06, "learning_rate": 0.00047477432296890674, "loss": 3.1913, "theoretical_loss": 4.361207839126308, "tokens_seen": 197853184 }, { "epoch": 0.06, "learning_rate": 0.0004747642928786359, "loss": 3.3772, "theoretical_loss": 4.361026808724247, "tokens_seen": 197918720 }, { "epoch": 0.06, "learning_rate": 0.0004747542627883651, "loss": 3.3336, "theoretical_loss": 4.360845855034049, "tokens_seen": 197984256 }, { "epoch": 0.06, "learning_rate": 0.0004747442326980943, "loss": 3.1099, "theoretical_loss": 4.360664977997828, "tokens_seen": 198049792 }, { "epoch": 0.06, "learning_rate": 0.00047473420260782346, "loss": 3.1798, "theoretical_loss": 4.360484177557766, "tokens_seen": 198115328 }, { "epoch": 0.06, "learning_rate": 0.0004747241725175527, "loss": 3.0723, "theoretical_loss": 4.360303453656103, "tokens_seen": 198180864 }, { "epoch": 0.06, "objective/train/docs_used": 122248, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.521057605743408, "objective/train/theoretical_loss": 4.360122806235145, "objective/train/tokens_used": 218706400, "theoretical_loss": 4.360122806235145, "tokens_seen": 198246400 }, { "epoch": 0.06, "learning_rate": 0.0004747141424272818, "loss": 3.3945, "theoretical_loss": 4.360122806235145, "tokens_seen": 198246400 }, { "epoch": 0.06, "learning_rate": 0.00047470411233701106, "loss": 3.0591, "theoretical_loss": 4.359942235237257, "tokens_seen": 198311936 }, { "epoch": 0.06, "learning_rate": 0.0004746940822467402, "loss": 2.9216, "theoretical_loss": 4.359761740604871, "tokens_seen": 198377472 }, { "epoch": 0.06, "learning_rate": 0.0004746840521564694, "loss": 3.3207, "theoretical_loss": 4.359581322280479, "tokens_seen": 198443008 }, { "epoch": 0.06, "learning_rate": 0.0004746740220661986, "loss": 3.3274, "theoretical_loss": 4.359400980206634, "tokens_seen": 198508544 }, { "epoch": 0.06, "learning_rate": 0.0004746639919759278, "loss": 3.3515, "theoretical_loss": 4.359220714325954, "tokens_seen": 198574080 }, { "epoch": 0.06, "learning_rate": 0.00047465396188565696, "loss": 3.272, "theoretical_loss": 4.359040524581116, "tokens_seen": 198639616 }, { "epoch": 0.06, "learning_rate": 0.0004746439317953862, "loss": 3.213, "theoretical_loss": 4.358860410914861, "tokens_seen": 198705152 }, { "epoch": 0.06, "learning_rate": 0.0004746339017051153, "loss": 3.1872, "theoretical_loss": 4.358680373269993, "tokens_seen": 198770688 }, { "epoch": 0.06, "learning_rate": 0.00047462387161484456, "loss": 2.9749, "theoretical_loss": 4.358500411589375, "tokens_seen": 198836224 }, { "epoch": 0.06, "learning_rate": 0.0004746138415245737, "loss": 3.2143, "theoretical_loss": 4.358320525815934, "tokens_seen": 198901760 }, { "epoch": 0.06, "learning_rate": 0.0004746038114343029, "loss": 3.2556, "theoretical_loss": 4.358140715892658, "tokens_seen": 198967296 }, { "epoch": 0.06, "learning_rate": 0.0004745937813440321, "loss": 3.2029, "theoretical_loss": 4.357960981762595, "tokens_seen": 199032832 }, { "epoch": 0.06, "learning_rate": 0.0004745837512537613, "loss": 3.1239, "theoretical_loss": 4.357781323368857, "tokens_seen": 199098368 }, { "epoch": 0.06, "learning_rate": 0.00047457372116349047, "loss": 2.9612, "theoretical_loss": 4.357601740654617, "tokens_seen": 199163904 }, { "epoch": 0.06, "learning_rate": 0.00047456369107321965, "loss": 3.0499, "theoretical_loss": 4.357422233563106, "tokens_seen": 199229440 }, { "epoch": 0.06, "learning_rate": 0.00047455366098294883, "loss": 3.0168, "theoretical_loss": 4.357242802037623, "tokens_seen": 199294976 }, { "epoch": 0.06, "learning_rate": 0.00047454363089267807, "loss": 3.276, "theoretical_loss": 4.35706344602152, "tokens_seen": 199360512 }, { "epoch": 0.06, "learning_rate": 0.0004745336008024072, "loss": 3.2373, "theoretical_loss": 4.356884165458217, "tokens_seen": 199426048 }, { "epoch": 0.06, "learning_rate": 0.00047452357071213643, "loss": 3.1589, "theoretical_loss": 4.356704960291191, "tokens_seen": 199491584 }, { "epoch": 0.06, "learning_rate": 0.00047451354062186555, "loss": 3.0203, "theoretical_loss": 4.35652583046398, "tokens_seen": 199557120 }, { "epoch": 0.06, "learning_rate": 0.0004745035105315948, "loss": 3.136, "theoretical_loss": 4.356346775920185, "tokens_seen": 199622656 }, { "epoch": 0.06, "learning_rate": 0.00047449348044132397, "loss": 3.2817, "theoretical_loss": 4.356167796603467, "tokens_seen": 199688192 }, { "epoch": 0.06, "learning_rate": 0.00047448345035105315, "loss": 3.282, "theoretical_loss": 4.355988892457546, "tokens_seen": 199753728 }, { "epoch": 0.06, "learning_rate": 0.0004744734202607824, "loss": 3.1863, "theoretical_loss": 4.355810063426204, "tokens_seen": 199819264 }, { "epoch": 0.06, "objective/train/docs_used": 123553, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0212910175323486, "objective/train/theoretical_loss": 4.355631309453283, "objective/train/tokens_used": 220344800, "theoretical_loss": 4.355631309453283, "tokens_seen": 199884800 }, { "epoch": 0.06, "learning_rate": 0.00047446339017051157, "loss": 3.171, "theoretical_loss": 4.355631309453283, "tokens_seen": 199884800 }, { "epoch": 0.06, "learning_rate": 0.00047445336008024075, "loss": 3.0987, "theoretical_loss": 4.355452630482685, "tokens_seen": 199950336 }, { "epoch": 0.06, "learning_rate": 0.00047444332998996993, "loss": 3.1614, "theoretical_loss": 4.355274026458375, "tokens_seen": 200015872 }, { "epoch": 0.06, "learning_rate": 0.0004744332998996991, "loss": 3.3795, "theoretical_loss": 4.355095497324373, "tokens_seen": 200081408 }, { "epoch": 0.06, "learning_rate": 0.0004744232698094283, "loss": 3.0216, "theoretical_loss": 4.354917043024765, "tokens_seen": 200146944 }, { "epoch": 0.06, "learning_rate": 0.00047441323971915753, "loss": 3.3482, "theoretical_loss": 4.354738663503692, "tokens_seen": 200212480 }, { "epoch": 0.06, "learning_rate": 0.00047440320962888666, "loss": 3.1463, "theoretical_loss": 4.354560358705358, "tokens_seen": 200278016 }, { "epoch": 0.06, "learning_rate": 0.0004743931795386159, "loss": 2.9351, "theoretical_loss": 4.354382128574027, "tokens_seen": 200343552 }, { "epoch": 0.06, "learning_rate": 0.000474383149448345, "loss": 3.3023, "theoretical_loss": 4.35420397305402, "tokens_seen": 200409088 }, { "epoch": 0.06, "learning_rate": 0.00047437311935807425, "loss": 3.2039, "theoretical_loss": 4.35402589208972, "tokens_seen": 200474624 }, { "epoch": 0.06, "learning_rate": 0.00047436308926780343, "loss": 3.1976, "theoretical_loss": 4.353847885625571, "tokens_seen": 200540160 }, { "epoch": 0.06, "learning_rate": 0.0004743530591775326, "loss": 3.0562, "theoretical_loss": 4.353669953606072, "tokens_seen": 200605696 }, { "epoch": 0.06, "learning_rate": 0.0004743430290872618, "loss": 3.4207, "theoretical_loss": 4.353492095975787, "tokens_seen": 200671232 }, { "epoch": 0.06, "learning_rate": 0.00047433299899699103, "loss": 3.0709, "theoretical_loss": 4.353314312679333, "tokens_seen": 200736768 }, { "epoch": 0.06, "learning_rate": 0.00047432296890672016, "loss": 3.1709, "theoretical_loss": 4.353136603661392, "tokens_seen": 200802304 }, { "epoch": 0.06, "learning_rate": 0.0004743129388164494, "loss": 3.2795, "theoretical_loss": 4.352958968866704, "tokens_seen": 200867840 }, { "epoch": 0.06, "learning_rate": 0.0004743029087261785, "loss": 3.3119, "theoretical_loss": 4.352781408240065, "tokens_seen": 200933376 }, { "epoch": 0.06, "learning_rate": 0.00047429287863590776, "loss": 3.0623, "theoretical_loss": 4.352603921726334, "tokens_seen": 200998912 }, { "epoch": 0.06, "learning_rate": 0.00047428284854563694, "loss": 3.4305, "theoretical_loss": 4.352426509270425, "tokens_seen": 201064448 }, { "epoch": 0.06, "learning_rate": 0.0004742728184553661, "loss": 3.1922, "theoretical_loss": 4.352249170817315, "tokens_seen": 201129984 }, { "epoch": 0.06, "learning_rate": 0.0004742627883650953, "loss": 3.3135, "theoretical_loss": 4.352071906312037, "tokens_seen": 201195520 }, { "epoch": 0.06, "learning_rate": 0.0004742527582748245, "loss": 3.1888, "theoretical_loss": 4.351894715699684, "tokens_seen": 201261056 }, { "epoch": 0.06, "learning_rate": 0.00047424272818455366, "loss": 3.4327, "theoretical_loss": 4.351717598925406, "tokens_seen": 201326592 }, { "epoch": 0.06, "learning_rate": 0.0004742326980942829, "loss": 3.3805, "theoretical_loss": 4.351540555934414, "tokens_seen": 201392128 }, { "epoch": 0.06, "learning_rate": 0.000474222668004012, "loss": 3.5804, "theoretical_loss": 4.351363586671976, "tokens_seen": 201457664 }, { "epoch": 0.06, "objective/train/docs_used": 124253, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.035176992416382, "objective/train/theoretical_loss": 4.351186691083417, "objective/train/tokens_used": 221983200, "theoretical_loss": 4.351186691083417, "tokens_seen": 201523200 }, { "epoch": 0.06, "learning_rate": 0.00047421263791374126, "loss": 3.126, "theoretical_loss": 4.351186691083417, "tokens_seen": 201523200 }, { "epoch": 0.06, "learning_rate": 0.0004742026078234704, "loss": 3.1474, "theoretical_loss": 4.351009869114124, "tokens_seen": 201588736 }, { "epoch": 0.06, "learning_rate": 0.0004741925777331996, "loss": 3.2053, "theoretical_loss": 4.350833120709539, "tokens_seen": 201654272 }, { "epoch": 0.06, "learning_rate": 0.0004741825476429288, "loss": 3.2585, "theoretical_loss": 4.350656445815164, "tokens_seen": 201719808 }, { "epoch": 0.06, "learning_rate": 0.000474172517552658, "loss": 3.1713, "theoretical_loss": 4.350479844376557, "tokens_seen": 201785344 }, { "epoch": 0.06, "learning_rate": 0.00047416248746238716, "loss": 3.3766, "theoretical_loss": 4.350303316339337, "tokens_seen": 201850880 }, { "epoch": 0.06, "learning_rate": 0.0004741524573721164, "loss": 3.2092, "theoretical_loss": 4.350126861649178, "tokens_seen": 201916416 }, { "epoch": 0.06, "learning_rate": 0.0004741424272818455, "loss": 3.1106, "theoretical_loss": 4.349950480251813, "tokens_seen": 201981952 }, { "epoch": 0.06, "learning_rate": 0.00047413239719157476, "loss": 3.0031, "theoretical_loss": 4.349774172093033, "tokens_seen": 202047488 }, { "epoch": 0.06, "learning_rate": 0.0004741223671013039, "loss": 3.4637, "theoretical_loss": 4.349597937118687, "tokens_seen": 202113024 }, { "epoch": 0.06, "learning_rate": 0.0004741123370110331, "loss": 3.169, "theoretical_loss": 4.3494217752746795, "tokens_seen": 202178560 }, { "epoch": 0.06, "learning_rate": 0.0004741023069207623, "loss": 3.2752, "theoretical_loss": 4.349245686506976, "tokens_seen": 202244096 }, { "epoch": 0.06, "learning_rate": 0.0004740922768304915, "loss": 3.3951, "theoretical_loss": 4.349069670761597, "tokens_seen": 202309632 }, { "epoch": 0.06, "learning_rate": 0.00047408224674022067, "loss": 3.12, "theoretical_loss": 4.348893727984619, "tokens_seen": 202375168 }, { "epoch": 0.06, "learning_rate": 0.00047407221664994985, "loss": 3.0903, "theoretical_loss": 4.348717858122178, "tokens_seen": 202440704 }, { "epoch": 0.06, "learning_rate": 0.00047406218655967903, "loss": 3.1313, "theoretical_loss": 4.348542061120469, "tokens_seen": 202506240 }, { "epoch": 0.06, "learning_rate": 0.00047405215646940827, "loss": 3.2476, "theoretical_loss": 4.348366336925739, "tokens_seen": 202571776 }, { "epoch": 0.06, "learning_rate": 0.0004740421263791374, "loss": 3.0185, "theoretical_loss": 4.3481906854842975, "tokens_seen": 202637312 }, { "epoch": 0.06, "learning_rate": 0.00047403209628886663, "loss": 3.1723, "theoretical_loss": 4.348015106742507, "tokens_seen": 202702848 }, { "epoch": 0.06, "learning_rate": 0.00047402206619859575, "loss": 3.1502, "theoretical_loss": 4.347839600646786, "tokens_seen": 202768384 }, { "epoch": 0.06, "learning_rate": 0.000474012036108325, "loss": 3.3803, "theoretical_loss": 4.347664167143615, "tokens_seen": 202833920 }, { "epoch": 0.06, "learning_rate": 0.00047400200601805417, "loss": 3.1256, "theoretical_loss": 4.347488806179528, "tokens_seen": 202899456 }, { "epoch": 0.06, "learning_rate": 0.00047399197592778335, "loss": 3.2037, "theoretical_loss": 4.347313517701114, "tokens_seen": 202964992 }, { "epoch": 0.06, "learning_rate": 0.00047398194583751253, "loss": 3.1465, "theoretical_loss": 4.347138301655021, "tokens_seen": 203030528 }, { "epoch": 0.06, "learning_rate": 0.00047397191574724177, "loss": 3.4567, "theoretical_loss": 4.346963157987954, "tokens_seen": 203096064 }, { "epoch": 0.06, "objective/train/docs_used": 125838, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.456173896789551, "objective/train/theoretical_loss": 4.346788086646671, "objective/train/tokens_used": 223621600, "theoretical_loss": 4.346788086646671, "tokens_seen": 203161600 }, { "epoch": 0.06, "learning_rate": 0.0004739618856569709, "loss": 3.2604, "theoretical_loss": 4.346788086646671, "tokens_seen": 203161600 }, { "epoch": 0.06, "learning_rate": 0.00047395185556670013, "loss": 3.2752, "theoretical_loss": 4.346613087577991, "tokens_seen": 203227136 }, { "epoch": 0.06, "learning_rate": 0.00047394182547642926, "loss": 3.2155, "theoretical_loss": 4.346438160728785, "tokens_seen": 203292672 }, { "epoch": 0.06, "learning_rate": 0.0004739317953861585, "loss": 3.2269, "theoretical_loss": 4.346263306045983, "tokens_seen": 203358208 }, { "epoch": 0.06, "learning_rate": 0.0004739217652958877, "loss": 3.1447, "theoretical_loss": 4.346088523476569, "tokens_seen": 203423744 }, { "epoch": 0.06, "learning_rate": 0.00047391173520561686, "loss": 3.1881, "theoretical_loss": 4.345913812967584, "tokens_seen": 203489280 }, { "epoch": 0.06, "learning_rate": 0.00047390170511534604, "loss": 3.148, "theoretical_loss": 4.345739174466127, "tokens_seen": 203554816 }, { "epoch": 0.06, "learning_rate": 0.0004738916750250752, "loss": 3.1253, "theoretical_loss": 4.345564607919348, "tokens_seen": 203620352 }, { "epoch": 0.06, "learning_rate": 0.0004738816449348044, "loss": 3.2427, "theoretical_loss": 4.3453901132744575, "tokens_seen": 203685888 }, { "epoch": 0.06, "learning_rate": 0.00047387161484453363, "loss": 3.1132, "theoretical_loss": 4.345215690478719, "tokens_seen": 203751424 }, { "epoch": 0.06, "learning_rate": 0.00047386158475426276, "loss": 3.0614, "theoretical_loss": 4.345041339479453, "tokens_seen": 203816960 }, { "epoch": 0.06, "learning_rate": 0.000473851554663992, "loss": 3.2345, "theoretical_loss": 4.3448670602240345, "tokens_seen": 203882496 }, { "epoch": 0.06, "learning_rate": 0.0004738415245737211, "loss": 2.9035, "theoretical_loss": 4.344692852659895, "tokens_seen": 203948032 }, { "epoch": 0.06, "learning_rate": 0.00047383149448345036, "loss": 3.2493, "theoretical_loss": 4.34451871673452, "tokens_seen": 204013568 }, { "epoch": 0.06, "learning_rate": 0.00047382146439317954, "loss": 3.2788, "theoretical_loss": 4.344344652395451, "tokens_seen": 204079104 }, { "epoch": 0.06, "learning_rate": 0.0004738114343029087, "loss": 3.0973, "theoretical_loss": 4.3441706595902865, "tokens_seen": 204144640 }, { "epoch": 0.06, "learning_rate": 0.0004738014042126379, "loss": 3.0732, "theoretical_loss": 4.343996738266677, "tokens_seen": 204210176 }, { "epoch": 0.06, "learning_rate": 0.00047379137412236714, "loss": 3.2518, "theoretical_loss": 4.343822888372331, "tokens_seen": 204275712 }, { "epoch": 0.06, "learning_rate": 0.00047378134403209626, "loss": 3.064, "theoretical_loss": 4.343649109855009, "tokens_seen": 204341248 }, { "epoch": 0.06, "learning_rate": 0.0004737713139418255, "loss": 3.3627, "theoretical_loss": 4.343475402662529, "tokens_seen": 204406784 }, { "epoch": 0.06, "learning_rate": 0.0004737612838515546, "loss": 3.0398, "theoretical_loss": 4.343301766742763, "tokens_seen": 204472320 }, { "epoch": 0.06, "learning_rate": 0.00047375125376128386, "loss": 3.0176, "theoretical_loss": 4.343128202043638, "tokens_seen": 204537856 }, { "epoch": 0.06, "learning_rate": 0.00047374122367101304, "loss": 3.257, "theoretical_loss": 4.342954708513136, "tokens_seen": 204603392 }, { "epoch": 0.06, "learning_rate": 0.0004737311935807422, "loss": 2.8876, "theoretical_loss": 4.342781286099291, "tokens_seen": 204668928 }, { "epoch": 0.06, "learning_rate": 0.00047372116349047146, "loss": 3.3289, "theoretical_loss": 4.3426079347501965, "tokens_seen": 204734464 }, { "epoch": 0.06, "objective/train/docs_used": 126453, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.122795820236206, "objective/train/theoretical_loss": 4.342434654413995, "objective/train/tokens_used": 225260000, "theoretical_loss": 4.342434654413995, "tokens_seen": 204800000 }, { "epoch": 0.06, "learning_rate": 0.0004737111334002006, "loss": 3.1005, "theoretical_loss": 4.342434654413995, "tokens_seen": 204800000 }, { "epoch": 0.06, "learning_rate": 0.0004737011033099298, "loss": 3.2419, "theoretical_loss": 4.342261445038888, "tokens_seen": 204865536 }, { "epoch": 0.06, "learning_rate": 0.000473691073219659, "loss": 3.2928, "theoretical_loss": 4.342088306573128, "tokens_seen": 204931072 }, { "epoch": 0.06, "learning_rate": 0.0004736810431293882, "loss": 2.9942, "theoretical_loss": 4.341915238965026, "tokens_seen": 204996608 }, { "epoch": 0.06, "learning_rate": 0.00047367101303911736, "loss": 3.3056, "theoretical_loss": 4.34174224216294, "tokens_seen": 205062144 }, { "epoch": 0.06, "learning_rate": 0.0004736609829488466, "loss": 3.3729, "theoretical_loss": 4.34156931611529, "tokens_seen": 205127680 }, { "epoch": 0.06, "learning_rate": 0.0004736509528585757, "loss": 3.1314, "theoretical_loss": 4.341396460770547, "tokens_seen": 205193216 }, { "epoch": 0.06, "learning_rate": 0.00047364092276830496, "loss": 3.0286, "theoretical_loss": 4.341223676077232, "tokens_seen": 205258752 }, { "epoch": 0.06, "learning_rate": 0.0004736308926780341, "loss": 3.3591, "theoretical_loss": 4.341050961983926, "tokens_seen": 205324288 }, { "epoch": 0.06, "learning_rate": 0.0004736208625877633, "loss": 3.2768, "theoretical_loss": 4.340878318439261, "tokens_seen": 205389824 }, { "epoch": 0.06, "learning_rate": 0.0004736108324974925, "loss": 3.1598, "theoretical_loss": 4.340705745391922, "tokens_seen": 205455360 }, { "epoch": 0.06, "learning_rate": 0.0004736008024072217, "loss": 3.0908, "theoretical_loss": 4.3405332427906504, "tokens_seen": 205520896 }, { "epoch": 0.06, "learning_rate": 0.00047359077231695087, "loss": 3.1934, "theoretical_loss": 4.340360810584238, "tokens_seen": 205586432 }, { "epoch": 0.06, "learning_rate": 0.00047358074222668005, "loss": 3.1539, "theoretical_loss": 4.340188448721532, "tokens_seen": 205651968 }, { "epoch": 0.06, "learning_rate": 0.00047357071213640923, "loss": 3.2624, "theoretical_loss": 4.3400161571514335, "tokens_seen": 205717504 }, { "epoch": 0.06, "learning_rate": 0.00047356068204613847, "loss": 3.2158, "theoretical_loss": 4.339843935822895, "tokens_seen": 205783040 }, { "epoch": 0.06, "learning_rate": 0.0004735506519558676, "loss": 3.1277, "theoretical_loss": 4.339671784684923, "tokens_seen": 205848576 }, { "epoch": 0.06, "learning_rate": 0.00047354062186559683, "loss": 3.0358, "theoretical_loss": 4.339499703686579, "tokens_seen": 205914112 }, { "epoch": 0.06, "learning_rate": 0.00047353059177532595, "loss": 3.2111, "theoretical_loss": 4.339327692776977, "tokens_seen": 205979648 }, { "epoch": 0.06, "learning_rate": 0.0004735205616850552, "loss": 3.1453, "theoretical_loss": 4.339155751905282, "tokens_seen": 206045184 }, { "epoch": 0.06, "learning_rate": 0.00047351053159478437, "loss": 3.2403, "theoretical_loss": 4.338983881020713, "tokens_seen": 206110720 }, { "epoch": 0.06, "learning_rate": 0.00047350050150451355, "loss": 3.3298, "theoretical_loss": 4.338812080072545, "tokens_seen": 206176256 }, { "epoch": 0.06, "learning_rate": 0.00047349047141424273, "loss": 3.3233, "theoretical_loss": 4.338640349010101, "tokens_seen": 206241792 }, { "epoch": 0.06, "learning_rate": 0.00047348044132397197, "loss": 3.2418, "theoretical_loss": 4.3384686877827585, "tokens_seen": 206307328 }, { "epoch": 0.06, "learning_rate": 0.0004734704112337011, "loss": 3.1538, "theoretical_loss": 4.338297096339951, "tokens_seen": 206372864 }, { "epoch": 0.06, "objective/train/docs_used": 127848, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0600967407226562, "objective/train/theoretical_loss": 4.33812557463116, "objective/train/tokens_used": 226898400, "theoretical_loss": 4.33812557463116, "tokens_seen": 206438400 }, { "epoch": 0.06, "learning_rate": 0.00047346038114343033, "loss": 3.0882, "theoretical_loss": 4.33812557463116, "tokens_seen": 206438400 }, { "epoch": 0.06, "learning_rate": 0.00047345035105315946, "loss": 3.2587, "theoretical_loss": 4.3379541226059235, "tokens_seen": 206503936 }, { "epoch": 0.06, "learning_rate": 0.0004734403209628887, "loss": 3.1858, "theoretical_loss": 4.337782740213827, "tokens_seen": 206569472 }, { "epoch": 0.06, "learning_rate": 0.0004734302908726179, "loss": 3.1159, "theoretical_loss": 4.337611427404514, "tokens_seen": 206635008 }, { "epoch": 0.06, "learning_rate": 0.00047342026078234706, "loss": 3.3164, "theoretical_loss": 4.337440184127679, "tokens_seen": 206700544 }, { "epoch": 0.06, "learning_rate": 0.00047341023069207624, "loss": 3.0572, "theoretical_loss": 4.337269010333065, "tokens_seen": 206766080 }, { "epoch": 0.06, "learning_rate": 0.0004734002006018054, "loss": 2.8251, "theoretical_loss": 4.337097905970471, "tokens_seen": 206831616 }, { "epoch": 0.06, "learning_rate": 0.0004733901705115346, "loss": 3.1861, "theoretical_loss": 4.336926870989748, "tokens_seen": 206897152 }, { "epoch": 0.06, "learning_rate": 0.00047338014042126383, "loss": 3.3199, "theoretical_loss": 4.336755905340797, "tokens_seen": 206962688 }, { "epoch": 0.06, "learning_rate": 0.00047337011033099296, "loss": 3.151, "theoretical_loss": 4.336585008973573, "tokens_seen": 207028224 }, { "epoch": 0.06, "learning_rate": 0.0004733600802407222, "loss": 3.0852, "theoretical_loss": 4.336414181838082, "tokens_seen": 207093760 }, { "epoch": 0.06, "learning_rate": 0.0004733500501504513, "loss": 3.1909, "theoretical_loss": 4.336243423884382, "tokens_seen": 207159296 }, { "epoch": 0.06, "learning_rate": 0.00047334002006018056, "loss": 3.2911, "theoretical_loss": 4.336072735062583, "tokens_seen": 207224832 }, { "epoch": 0.06, "learning_rate": 0.00047332998996990974, "loss": 3.2248, "theoretical_loss": 4.335902115322847, "tokens_seen": 207290368 }, { "epoch": 0.06, "learning_rate": 0.0004733199598796389, "loss": 2.9551, "theoretical_loss": 4.335731564615387, "tokens_seen": 207355904 }, { "epoch": 0.06, "learning_rate": 0.0004733099297893681, "loss": 3.3696, "theoretical_loss": 4.335561082890468, "tokens_seen": 207421440 }, { "epoch": 0.06, "learning_rate": 0.00047329989969909734, "loss": 3.4282, "theoretical_loss": 4.335390670098407, "tokens_seen": 207486976 }, { "epoch": 0.06, "learning_rate": 0.00047328986960882646, "loss": 3.2705, "theoretical_loss": 4.335220326189571, "tokens_seen": 207552512 }, { "epoch": 0.06, "learning_rate": 0.0004732798395185557, "loss": 3.205, "theoretical_loss": 4.335050051114379, "tokens_seen": 207618048 }, { "epoch": 0.06, "learning_rate": 0.0004732698094282848, "loss": 2.9493, "theoretical_loss": 4.334879844823304, "tokens_seen": 207683584 }, { "epoch": 0.06, "learning_rate": 0.00047325977933801406, "loss": 3.0003, "theoretical_loss": 4.334709707266865, "tokens_seen": 207749120 }, { "epoch": 0.06, "learning_rate": 0.00047324974924774324, "loss": 3.1429, "theoretical_loss": 4.334539638395636, "tokens_seen": 207814656 }, { "epoch": 0.06, "learning_rate": 0.0004732397191574724, "loss": 3.2752, "theoretical_loss": 4.334369638160242, "tokens_seen": 207880192 }, { "epoch": 0.06, "learning_rate": 0.0004732296890672016, "loss": 3.1693, "theoretical_loss": 4.334199706511358, "tokens_seen": 207945728 }, { "epoch": 0.06, "learning_rate": 0.0004732196589769308, "loss": 3.402, "theoretical_loss": 4.334029843399709, "tokens_seen": 208011264 }, { "epoch": 0.06, "objective/train/docs_used": 128338, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.665210247039795, "objective/train/theoretical_loss": 4.333860048776074, "objective/train/tokens_used": 228536800, "theoretical_loss": 4.333860048776074, "tokens_seen": 208076800 }, { "epoch": 0.06, "learning_rate": 0.00047320962888665997, "loss": 3.2236, "theoretical_loss": 4.333860048776074, "tokens_seen": 208076800 }, { "epoch": 0.06, "learning_rate": 0.0004731995987963892, "loss": 3.167, "theoretical_loss": 4.33369032259128, "tokens_seen": 208142336 }, { "epoch": 0.06, "learning_rate": 0.00047318956870611833, "loss": 3.1083, "theoretical_loss": 4.333520664796206, "tokens_seen": 208207872 }, { "epoch": 0.06, "learning_rate": 0.00047317953861584757, "loss": 3.1765, "theoretical_loss": 4.33335107534178, "tokens_seen": 208273408 }, { "epoch": 0.06, "learning_rate": 0.00047316950852557675, "loss": 3.1977, "theoretical_loss": 4.333181554178985, "tokens_seen": 208338944 }, { "epoch": 0.06, "learning_rate": 0.00047315947843530593, "loss": 2.9084, "theoretical_loss": 4.3330121012588485, "tokens_seen": 208404480 }, { "epoch": 0.06, "learning_rate": 0.0004731494483450351, "loss": 3.4184, "theoretical_loss": 4.332842716532454, "tokens_seen": 208470016 }, { "epoch": 0.06, "learning_rate": 0.0004731394182547643, "loss": 3.2963, "theoretical_loss": 4.332673399950932, "tokens_seen": 208535552 }, { "epoch": 0.06, "learning_rate": 0.00047312938816449347, "loss": 3.052, "theoretical_loss": 4.332504151465464, "tokens_seen": 208601088 }, { "epoch": 0.06, "learning_rate": 0.0004731193580742227, "loss": 3.0663, "theoretical_loss": 4.332334971027284, "tokens_seen": 208666624 }, { "epoch": 0.06, "learning_rate": 0.00047310932798395183, "loss": 3.0481, "theoretical_loss": 4.332165858587672, "tokens_seen": 208732160 }, { "epoch": 0.06, "learning_rate": 0.00047309929789368107, "loss": 3.3365, "theoretical_loss": 4.331996814097963, "tokens_seen": 208797696 }, { "epoch": 0.06, "learning_rate": 0.0004730892678034102, "loss": 3.1267, "theoretical_loss": 4.331827837509538, "tokens_seen": 208863232 }, { "epoch": 0.06, "learning_rate": 0.00047307923771313943, "loss": 3.3878, "theoretical_loss": 4.331658928773831, "tokens_seen": 208928768 }, { "epoch": 0.06, "learning_rate": 0.0004730692076228686, "loss": 3.0899, "theoretical_loss": 4.331490087842324, "tokens_seen": 208994304 }, { "epoch": 0.06, "learning_rate": 0.0004730591775325978, "loss": 3.2895, "theoretical_loss": 4.33132131466655, "tokens_seen": 209059840 }, { "epoch": 0.06, "learning_rate": 0.000473049147442327, "loss": 3.2144, "theoretical_loss": 4.3311526091980905, "tokens_seen": 209125376 }, { "epoch": 0.06, "learning_rate": 0.00047303911735205615, "loss": 2.7723, "theoretical_loss": 4.330983971388578, "tokens_seen": 209190912 }, { "epoch": 0.06, "learning_rate": 0.00047302908726178534, "loss": 2.9338, "theoretical_loss": 4.330815401189695, "tokens_seen": 209256448 }, { "epoch": 0.06, "learning_rate": 0.00047301905717151457, "loss": 3.1443, "theoretical_loss": 4.330646898553173, "tokens_seen": 209321984 }, { "epoch": 0.06, "learning_rate": 0.0004730090270812437, "loss": 3.2174, "theoretical_loss": 4.330478463430792, "tokens_seen": 209387520 }, { "epoch": 0.06, "learning_rate": 0.00047299899699097293, "loss": 3.2128, "theoretical_loss": 4.330310095774383, "tokens_seen": 209453056 }, { "epoch": 0.06, "learning_rate": 0.0004729889669007021, "loss": 3.1469, "theoretical_loss": 4.330141795535828, "tokens_seen": 209518592 }, { "epoch": 0.06, "learning_rate": 0.0004729789368104313, "loss": 3.2627, "theoretical_loss": 4.329973562667053, "tokens_seen": 209584128 }, { "epoch": 0.06, "learning_rate": 0.00047296890672016053, "loss": 3.3848, "theoretical_loss": 4.3298053971200385, "tokens_seen": 209649664 }, { "epoch": 0.06, "objective/train/docs_used": 129483, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.797501802444458, "objective/train/theoretical_loss": 4.329637298846812, "objective/train/tokens_used": 230175200, "theoretical_loss": 4.329637298846812, "tokens_seen": 209715200 }, { "epoch": 0.06, "learning_rate": 0.00047295887662988966, "loss": 3.3015, "theoretical_loss": 4.329637298846812, "tokens_seen": 209715200 }, { "epoch": 0.06, "learning_rate": 0.0004729488465396189, "loss": 3.2439, "theoretical_loss": 4.329469267799451, "tokens_seen": 209780736 }, { "epoch": 0.06, "learning_rate": 0.0004729388164493481, "loss": 3.2575, "theoretical_loss": 4.32930130393008, "tokens_seen": 209846272 }, { "epoch": 0.06, "learning_rate": 0.00047292878635907726, "loss": 3.1579, "theoretical_loss": 4.329133407190876, "tokens_seen": 209911808 }, { "epoch": 0.06, "learning_rate": 0.00047291875626880644, "loss": 3.204, "theoretical_loss": 4.3289655775340625, "tokens_seen": 209977344 }, { "epoch": 0.06, "learning_rate": 0.0004729087261785356, "loss": 3.1764, "theoretical_loss": 4.328797814911912, "tokens_seen": 210042880 }, { "epoch": 0.06, "learning_rate": 0.0004728986960882648, "loss": 3.5043, "theoretical_loss": 4.328630119276747, "tokens_seen": 210108416 }, { "epoch": 0.06, "learning_rate": 0.00047288866599799403, "loss": 3.3447, "theoretical_loss": 4.328462490580938, "tokens_seen": 210173952 }, { "epoch": 0.06, "learning_rate": 0.00047287863590772316, "loss": 3.1687, "theoretical_loss": 4.328294928776903, "tokens_seen": 210239488 }, { "epoch": 0.06, "learning_rate": 0.0004728686058174524, "loss": 3.3405, "theoretical_loss": 4.328127433817112, "tokens_seen": 210305024 }, { "epoch": 0.06, "learning_rate": 0.0004728585757271815, "loss": 3.0517, "theoretical_loss": 4.327960005654081, "tokens_seen": 210370560 }, { "epoch": 0.06, "learning_rate": 0.00047284854563691076, "loss": 3.3842, "theoretical_loss": 4.327792644240374, "tokens_seen": 210436096 }, { "epoch": 0.06, "learning_rate": 0.00047283851554663994, "loss": 3.2548, "theoretical_loss": 4.327625349528605, "tokens_seen": 210501632 }, { "epoch": 0.06, "learning_rate": 0.0004728284854563691, "loss": 3.3037, "theoretical_loss": 4.327458121471436, "tokens_seen": 210567168 }, { "epoch": 0.06, "learning_rate": 0.0004728184553660983, "loss": 3.1623, "theoretical_loss": 4.3272909600215765, "tokens_seen": 210632704 }, { "epoch": 0.06, "learning_rate": 0.00047280842527582754, "loss": 2.8898, "theoretical_loss": 4.327123865131786, "tokens_seen": 210698240 }, { "epoch": 0.06, "learning_rate": 0.00047279839518555666, "loss": 3.2277, "theoretical_loss": 4.326956836754871, "tokens_seen": 210763776 }, { "epoch": 0.06, "learning_rate": 0.0004727883650952859, "loss": 3.2198, "theoretical_loss": 4.326789874843685, "tokens_seen": 210829312 }, { "epoch": 0.06, "learning_rate": 0.000472778335005015, "loss": 3.577, "theoretical_loss": 4.326622979351132, "tokens_seen": 210894848 }, { "epoch": 0.06, "learning_rate": 0.00047276830491474426, "loss": 3.1072, "theoretical_loss": 4.326456150230163, "tokens_seen": 210960384 }, { "epoch": 0.06, "learning_rate": 0.00047275827482447344, "loss": 3.3812, "theoretical_loss": 4.326289387433776, "tokens_seen": 211025920 }, { "epoch": 0.06, "learning_rate": 0.0004727482447342026, "loss": 3.3247, "theoretical_loss": 4.326122690915017, "tokens_seen": 211091456 }, { "epoch": 0.06, "learning_rate": 0.0004727382146439318, "loss": 3.1334, "theoretical_loss": 4.325956060626982, "tokens_seen": 211156992 }, { "epoch": 0.06, "learning_rate": 0.000472728184553661, "loss": 3.2365, "theoretical_loss": 4.325789496522812, "tokens_seen": 211222528 }, { "epoch": 0.06, "learning_rate": 0.00047271815446339017, "loss": 3.3295, "theoretical_loss": 4.325622998555697, "tokens_seen": 211288064 }, { "epoch": 0.06, "objective/train/docs_used": 130263, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.355026960372925, "objective/train/theoretical_loss": 4.3254565666788745, "objective/train/tokens_used": 231813600, "theoretical_loss": 4.3254565666788745, "tokens_seen": 211353600 }, { "epoch": 0.06, "learning_rate": 0.0004727081243731194, "loss": 3.3001, "theoretical_loss": 4.3254565666788745, "tokens_seen": 211353600 }, { "epoch": 0.06, "learning_rate": 0.00047269809428284853, "loss": 3.2941, "theoretical_loss": 4.325290200845629, "tokens_seen": 211419136 }, { "epoch": 0.06, "learning_rate": 0.00047268806419257777, "loss": 3.3354, "theoretical_loss": 4.3251239010092934, "tokens_seen": 211484672 }, { "epoch": 0.06, "learning_rate": 0.00047267803410230695, "loss": 3.3392, "theoretical_loss": 4.324957667123249, "tokens_seen": 211550208 }, { "epoch": 0.06, "learning_rate": 0.00047266800401203613, "loss": 3.1309, "theoretical_loss": 4.32479149914092, "tokens_seen": 211615744 }, { "epoch": 0.06, "learning_rate": 0.0004726579739217653, "loss": 3.2534, "theoretical_loss": 4.324625397015783, "tokens_seen": 211681280 }, { "epoch": 0.06, "learning_rate": 0.0004726479438314945, "loss": 3.0925, "theoretical_loss": 4.3244593607013595, "tokens_seen": 211746816 }, { "epoch": 0.06, "learning_rate": 0.00047263791374122367, "loss": 3.2174, "theoretical_loss": 4.324293390151218, "tokens_seen": 211812352 }, { "epoch": 0.06, "learning_rate": 0.0004726278836509529, "loss": 3.1953, "theoretical_loss": 4.324127485318975, "tokens_seen": 211877888 }, { "epoch": 0.06, "learning_rate": 0.00047261785356068203, "loss": 3.0756, "theoretical_loss": 4.323961646158294, "tokens_seen": 211943424 }, { "epoch": 0.06, "learning_rate": 0.00047260782347041127, "loss": 3.3479, "theoretical_loss": 4.323795872622884, "tokens_seen": 212008960 }, { "epoch": 0.06, "learning_rate": 0.0004725977933801404, "loss": 3.1166, "theoretical_loss": 4.323630164666502, "tokens_seen": 212074496 }, { "epoch": 0.06, "learning_rate": 0.00047258776328986963, "loss": 3.3872, "theoretical_loss": 4.323464522242954, "tokens_seen": 212140032 }, { "epoch": 0.06, "learning_rate": 0.0004725777331995988, "loss": 3.0378, "theoretical_loss": 4.323298945306089, "tokens_seen": 212205568 }, { "epoch": 0.06, "learning_rate": 0.000472567703109328, "loss": 2.9616, "theoretical_loss": 4.3231334338098035, "tokens_seen": 212271104 }, { "epoch": 0.06, "learning_rate": 0.0004725576730190572, "loss": 3.1819, "theoretical_loss": 4.322967987708043, "tokens_seen": 212336640 }, { "epoch": 0.06, "learning_rate": 0.00047254764292878636, "loss": 3.2994, "theoretical_loss": 4.322802606954798, "tokens_seen": 212402176 }, { "epoch": 0.06, "learning_rate": 0.00047253761283851554, "loss": 3.2778, "theoretical_loss": 4.322637291504106, "tokens_seen": 212467712 }, { "epoch": 0.06, "learning_rate": 0.00047252758274824477, "loss": 3.3196, "theoretical_loss": 4.32247204131005, "tokens_seen": 212533248 }, { "epoch": 0.06, "learning_rate": 0.0004725175526579739, "loss": 3.2477, "theoretical_loss": 4.322306856326761, "tokens_seen": 212598784 }, { "epoch": 0.06, "learning_rate": 0.00047250752256770313, "loss": 3.4815, "theoretical_loss": 4.322141736508415, "tokens_seen": 212664320 }, { "epoch": 0.06, "learning_rate": 0.0004724974924774323, "loss": 3.3946, "theoretical_loss": 4.321976681809236, "tokens_seen": 212729856 }, { "epoch": 0.06, "learning_rate": 0.0004724874623871615, "loss": 3.2502, "theoretical_loss": 4.321811692183491, "tokens_seen": 212795392 }, { "epoch": 0.06, "learning_rate": 0.0004724774322968907, "loss": 3.2951, "theoretical_loss": 4.321646767585497, "tokens_seen": 212860928 }, { "epoch": 0.06, "learning_rate": 0.00047246740220661986, "loss": 3.0522, "theoretical_loss": 4.3214819079696145, "tokens_seen": 212926464 }, { "epoch": 0.06, "objective/train/docs_used": 131630, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6950013637542725, "objective/train/theoretical_loss": 4.321317113290252, "objective/train/tokens_used": 233452000, "theoretical_loss": 4.321317113290252, "tokens_seen": 212992000 }, { "epoch": 0.06, "learning_rate": 0.00047245737211634904, "loss": 3.3024, "theoretical_loss": 4.321317113290252, "tokens_seen": 212992000 }, { "epoch": 0.06, "learning_rate": 0.0004724473420260783, "loss": 3.0431, "theoretical_loss": 4.321152383501863, "tokens_seen": 213057536 }, { "epoch": 0.06, "learning_rate": 0.0004724373119358074, "loss": 2.8804, "theoretical_loss": 4.320987718558945, "tokens_seen": 213123072 }, { "epoch": 0.06, "learning_rate": 0.00047242728184553664, "loss": 3.1601, "theoretical_loss": 4.320823118416046, "tokens_seen": 213188608 }, { "epoch": 0.06, "learning_rate": 0.00047241725175526576, "loss": 3.385, "theoretical_loss": 4.320658583027755, "tokens_seen": 213254144 }, { "epoch": 0.06, "learning_rate": 0.000472407221664995, "loss": 3.2701, "theoretical_loss": 4.32049411234871, "tokens_seen": 213319680 }, { "epoch": 0.06, "learning_rate": 0.0004723971915747242, "loss": 3.1385, "theoretical_loss": 4.3203297063335935, "tokens_seen": 213385216 }, { "epoch": 0.06, "learning_rate": 0.00047238716148445336, "loss": 3.1332, "theoretical_loss": 4.320165364937134, "tokens_seen": 213450752 }, { "epoch": 0.06, "learning_rate": 0.00047237713139418254, "loss": 3.3959, "theoretical_loss": 4.320001088114105, "tokens_seen": 213516288 }, { "epoch": 0.06, "learning_rate": 0.0004723671013039117, "loss": 3.5302, "theoretical_loss": 4.319836875819325, "tokens_seen": 213581824 }, { "epoch": 0.06, "learning_rate": 0.0004723570712136409, "loss": 3.3552, "theoretical_loss": 4.31967272800766, "tokens_seen": 213647360 }, { "epoch": 0.06, "learning_rate": 0.00047234704112337014, "loss": 3.2224, "theoretical_loss": 4.319508644634021, "tokens_seen": 213712896 }, { "epoch": 0.06, "learning_rate": 0.00047233701103309927, "loss": 3.1956, "theoretical_loss": 4.319344625653361, "tokens_seen": 213778432 }, { "epoch": 0.06, "learning_rate": 0.0004723269809428285, "loss": 3.0912, "theoretical_loss": 4.319180671020684, "tokens_seen": 213843968 }, { "epoch": 0.06, "learning_rate": 0.0004723169508525577, "loss": 3.22, "theoretical_loss": 4.319016780691033, "tokens_seen": 213909504 }, { "epoch": 0.06, "learning_rate": 0.00047230692076228686, "loss": 3.2483, "theoretical_loss": 4.318852954619501, "tokens_seen": 213975040 }, { "epoch": 0.06, "learning_rate": 0.00047229689067201605, "loss": 3.19, "theoretical_loss": 4.318689192761225, "tokens_seen": 214040576 }, { "epoch": 0.06, "learning_rate": 0.0004722868605817452, "loss": 3.0916, "theoretical_loss": 4.318525495071385, "tokens_seen": 214106112 }, { "epoch": 0.06, "learning_rate": 0.0004722768304914744, "loss": 3.5612, "theoretical_loss": 4.318361861505207, "tokens_seen": 214171648 }, { "epoch": 0.06, "learning_rate": 0.00047226680040120364, "loss": 3.2789, "theoretical_loss": 4.318198292017964, "tokens_seen": 214237184 }, { "epoch": 0.06, "learning_rate": 0.00047225677031093277, "loss": 3.219, "theoretical_loss": 4.318034786564971, "tokens_seen": 214302720 }, { "epoch": 0.06, "learning_rate": 0.000472246740220662, "loss": 3.3485, "theoretical_loss": 4.31787134510159, "tokens_seen": 214368256 }, { "epoch": 0.06, "learning_rate": 0.00047223671013039113, "loss": 3.2924, "theoretical_loss": 4.3177079675832255, "tokens_seen": 214433792 }, { "epoch": 0.06, "learning_rate": 0.00047222668004012037, "loss": 3.2976, "theoretical_loss": 4.317544653965329, "tokens_seen": 214499328 }, { "epoch": 0.07, "learning_rate": 0.0004722166499498496, "loss": 3.2069, "theoretical_loss": 4.3173814042033944, "tokens_seen": 214564864 }, { "epoch": 0.07, "objective/train/docs_used": 132228, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7650396823883057, "objective/train/theoretical_loss": 4.317218218252963, "objective/train/tokens_used": 235090400, "theoretical_loss": 4.317218218252963, "tokens_seen": 214630400 }, { "epoch": 0.07, "learning_rate": 0.00047220661985957873, "loss": 3.3818, "theoretical_loss": 4.317218218252963, "tokens_seen": 214630400 }, { "epoch": 0.07, "learning_rate": 0.00047219658976930797, "loss": 3.1204, "theoretical_loss": 4.317055096069618, "tokens_seen": 214695936 }, { "epoch": 0.07, "learning_rate": 0.00047218655967903715, "loss": 3.2258, "theoretical_loss": 4.316892037608987, "tokens_seen": 214761472 }, { "epoch": 0.07, "learning_rate": 0.00047217652958876633, "loss": 2.9512, "theoretical_loss": 4.316729042826745, "tokens_seen": 214827008 }, { "epoch": 0.07, "learning_rate": 0.0004721664994984955, "loss": 3.3851, "theoretical_loss": 4.316566111678609, "tokens_seen": 214892544 }, { "epoch": 0.07, "learning_rate": 0.0004721564694082247, "loss": 3.307, "theoretical_loss": 4.316403244120339, "tokens_seen": 214958080 }, { "epoch": 0.07, "learning_rate": 0.00047214643931795387, "loss": 3.0003, "theoretical_loss": 4.3162404401077445, "tokens_seen": 215023616 }, { "epoch": 0.07, "learning_rate": 0.0004721364092276831, "loss": 3.4895, "theoretical_loss": 4.316077699596671, "tokens_seen": 215089152 }, { "epoch": 0.07, "learning_rate": 0.00047212637913741223, "loss": 3.1155, "theoretical_loss": 4.315915022543016, "tokens_seen": 215154688 }, { "epoch": 0.07, "learning_rate": 0.00047211634904714147, "loss": 3.0629, "theoretical_loss": 4.315752408902716, "tokens_seen": 215220224 }, { "epoch": 0.07, "learning_rate": 0.0004721063189568706, "loss": 3.2205, "theoretical_loss": 4.315589858631755, "tokens_seen": 215285760 }, { "epoch": 0.07, "learning_rate": 0.00047209628886659983, "loss": 3.2874, "theoretical_loss": 4.315427371686157, "tokens_seen": 215351296 }, { "epoch": 0.07, "learning_rate": 0.000472086258776329, "loss": 3.3042, "theoretical_loss": 4.315264948021994, "tokens_seen": 215416832 }, { "epoch": 0.07, "learning_rate": 0.0004720762286860582, "loss": 3.3027, "theoretical_loss": 4.315102587595379, "tokens_seen": 215482368 }, { "epoch": 0.07, "learning_rate": 0.0004720661985957874, "loss": 3.1877, "theoretical_loss": 4.31494029036247, "tokens_seen": 215547904 }, { "epoch": 0.07, "learning_rate": 0.00047205616850551656, "loss": 3.2381, "theoretical_loss": 4.314778056279468, "tokens_seen": 215613440 }, { "epoch": 0.07, "learning_rate": 0.00047204613841524574, "loss": 3.0961, "theoretical_loss": 4.314615885302619, "tokens_seen": 215678976 }, { "epoch": 0.07, "learning_rate": 0.00047203610832497497, "loss": 3.0676, "theoretical_loss": 4.314453777388209, "tokens_seen": 215744512 }, { "epoch": 0.07, "learning_rate": 0.0004720260782347041, "loss": 3.2471, "theoretical_loss": 4.314291732492573, "tokens_seen": 215810048 }, { "epoch": 0.07, "learning_rate": 0.00047201604814443333, "loss": 3.2031, "theoretical_loss": 4.314129750572087, "tokens_seen": 215875584 }, { "epoch": 0.07, "learning_rate": 0.0004720060180541625, "loss": 2.8863, "theoretical_loss": 4.3139678315831675, "tokens_seen": 215941120 }, { "epoch": 0.07, "learning_rate": 0.0004719959879638917, "loss": 3.0952, "theoretical_loss": 4.313805975482278, "tokens_seen": 216006656 }, { "epoch": 0.07, "learning_rate": 0.0004719859578736209, "loss": 3.196, "theoretical_loss": 4.313644182225926, "tokens_seen": 216072192 }, { "epoch": 0.07, "learning_rate": 0.00047197592778335006, "loss": 2.9889, "theoretical_loss": 4.313482451770659, "tokens_seen": 216137728 }, { "epoch": 0.07, "learning_rate": 0.00047196589769307924, "loss": 3.1137, "theoretical_loss": 4.313320784073069, "tokens_seen": 216203264 }, { "epoch": 0.07, "objective/train/docs_used": 133529, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.224602460861206, "objective/train/theoretical_loss": 4.3131591790897925, "objective/train/tokens_used": 236728800, "theoretical_loss": 4.3131591790897925, "tokens_seen": 216268800 }, { "epoch": 0.07, "learning_rate": 0.0004719558676028085, "loss": 3.3034, "theoretical_loss": 4.3131591790897925, "tokens_seen": 216268800 }, { "epoch": 0.07, "learning_rate": 0.0004719458375125376, "loss": 3.0825, "theoretical_loss": 4.3129976367775065, "tokens_seen": 216334336 }, { "epoch": 0.07, "learning_rate": 0.00047193580742226684, "loss": 3.293, "theoretical_loss": 4.312836157092934, "tokens_seen": 216399872 }, { "epoch": 0.07, "learning_rate": 0.00047192577733199596, "loss": 3.172, "theoretical_loss": 4.312674739992839, "tokens_seen": 216465408 }, { "epoch": 0.07, "learning_rate": 0.0004719157472417252, "loss": 3.128, "theoretical_loss": 4.31251338543403, "tokens_seen": 216530944 }, { "epoch": 0.07, "learning_rate": 0.0004719057171514544, "loss": 3.3622, "theoretical_loss": 4.312352093373354, "tokens_seen": 216596480 }, { "epoch": 0.07, "learning_rate": 0.00047189568706118356, "loss": 3.3549, "theoretical_loss": 4.312190863767708, "tokens_seen": 216662016 }, { "epoch": 0.07, "learning_rate": 0.00047188565697091274, "loss": 3.1291, "theoretical_loss": 4.312029696574027, "tokens_seen": 216727552 }, { "epoch": 0.07, "learning_rate": 0.0004718756268806419, "loss": 2.8499, "theoretical_loss": 4.311868591749287, "tokens_seen": 216793088 }, { "epoch": 0.07, "learning_rate": 0.0004718655967903711, "loss": 3.0329, "theoretical_loss": 4.311707549250514, "tokens_seen": 216858624 }, { "epoch": 0.07, "learning_rate": 0.00047185556670010034, "loss": 3.5888, "theoretical_loss": 4.311546569034767, "tokens_seen": 216924160 }, { "epoch": 0.07, "learning_rate": 0.00047184553660982947, "loss": 3.4006, "theoretical_loss": 4.311385651059155, "tokens_seen": 216989696 }, { "epoch": 0.07, "learning_rate": 0.0004718355065195587, "loss": 3.2021, "theoretical_loss": 4.311224795280825, "tokens_seen": 217055232 }, { "epoch": 0.07, "learning_rate": 0.0004718254764292879, "loss": 3.2574, "theoretical_loss": 4.3110640016569715, "tokens_seen": 217120768 }, { "epoch": 0.07, "learning_rate": 0.00047181544633901706, "loss": 3.4922, "theoretical_loss": 4.310903270144825, "tokens_seen": 217186304 }, { "epoch": 0.07, "learning_rate": 0.00047180541624874625, "loss": 3.4122, "theoretical_loss": 4.310742600701664, "tokens_seen": 217251840 }, { "epoch": 0.07, "learning_rate": 0.0004717953861584754, "loss": 2.913, "theoretical_loss": 4.310581993284805, "tokens_seen": 217317376 }, { "epoch": 0.07, "learning_rate": 0.0004717853560682046, "loss": 3.0789, "theoretical_loss": 4.310421447851609, "tokens_seen": 217382912 }, { "epoch": 0.07, "learning_rate": 0.00047177532597793384, "loss": 3.1302, "theoretical_loss": 4.310260964359479, "tokens_seen": 217448448 }, { "epoch": 0.07, "learning_rate": 0.00047176529588766297, "loss": 3.0638, "theoretical_loss": 4.310100542765858, "tokens_seen": 217513984 }, { "epoch": 0.07, "learning_rate": 0.0004717552657973922, "loss": 3.1546, "theoretical_loss": 4.309940183028236, "tokens_seen": 217579520 }, { "epoch": 0.07, "learning_rate": 0.00047174523570712133, "loss": 3.3793, "theoretical_loss": 4.309779885104139, "tokens_seen": 217645056 }, { "epoch": 0.07, "learning_rate": 0.00047173520561685057, "loss": 3.2103, "theoretical_loss": 4.309619648951139, "tokens_seen": 217710592 }, { "epoch": 0.07, "learning_rate": 0.00047172517552657975, "loss": 3.3218, "theoretical_loss": 4.3094594745268475, "tokens_seen": 217776128 }, { "epoch": 0.07, "learning_rate": 0.00047171514543630893, "loss": 3.0378, "theoretical_loss": 4.30929936178892, "tokens_seen": 217841664 }, { "epoch": 0.07, "objective/train/docs_used": 134228, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2407047748565674, "objective/train/theoretical_loss": 4.309139310695053, "objective/train/tokens_used": 238367200, "theoretical_loss": 4.309139310695053, "tokens_seen": 217907200 }, { "epoch": 0.07, "learning_rate": 0.0004717051153460381, "loss": 3.2923, "theoretical_loss": 4.309139310695053, "tokens_seen": 217907200 }, { "epoch": 0.07, "learning_rate": 0.00047169508525576735, "loss": 3.2057, "theoretical_loss": 4.308979321202983, "tokens_seen": 217972736 }, { "epoch": 0.07, "learning_rate": 0.0004716850551654965, "loss": 3.2015, "theoretical_loss": 4.308819393270491, "tokens_seen": 218038272 }, { "epoch": 0.07, "learning_rate": 0.0004716750250752257, "loss": 3.064, "theoretical_loss": 4.308659526855396, "tokens_seen": 218103808 }, { "epoch": 0.07, "learning_rate": 0.00047166499498495484, "loss": 3.0323, "theoretical_loss": 4.308499721915563, "tokens_seen": 218169344 }, { "epoch": 0.07, "learning_rate": 0.00047165496489468407, "loss": 3.2744, "theoretical_loss": 4.308339978408897, "tokens_seen": 218234880 }, { "epoch": 0.07, "learning_rate": 0.00047164493480441325, "loss": 3.1236, "theoretical_loss": 4.308180296293341, "tokens_seen": 218300416 }, { "epoch": 0.07, "learning_rate": 0.00047163490471414243, "loss": 3.117, "theoretical_loss": 4.308020675526883, "tokens_seen": 218365952 }, { "epoch": 0.07, "learning_rate": 0.0004716248746238716, "loss": 3.1281, "theoretical_loss": 4.307861116067554, "tokens_seen": 218431488 }, { "epoch": 0.07, "learning_rate": 0.0004716148445336008, "loss": 2.9074, "theoretical_loss": 4.30770161787342, "tokens_seen": 218497024 }, { "epoch": 0.07, "learning_rate": 0.00047160481444333, "loss": 3.2527, "theoretical_loss": 4.307542180902594, "tokens_seen": 218562560 }, { "epoch": 0.07, "learning_rate": 0.0004715947843530592, "loss": 3.0822, "theoretical_loss": 4.307382805113228, "tokens_seen": 218628096 }, { "epoch": 0.07, "learning_rate": 0.00047158475426278834, "loss": 3.3932, "theoretical_loss": 4.307223490463516, "tokens_seen": 218693632 }, { "epoch": 0.07, "learning_rate": 0.0004715747241725176, "loss": 3.1793, "theoretical_loss": 4.307064236911692, "tokens_seen": 218759168 }, { "epoch": 0.07, "learning_rate": 0.0004715646940822467, "loss": 3.3369, "theoretical_loss": 4.30690504441603, "tokens_seen": 218824704 }, { "epoch": 0.07, "learning_rate": 0.00047155466399197594, "loss": 3.1318, "theoretical_loss": 4.306745912934849, "tokens_seen": 218890240 }, { "epoch": 0.07, "learning_rate": 0.0004715446339017051, "loss": 3.0594, "theoretical_loss": 4.306586842426504, "tokens_seen": 218955776 }, { "epoch": 0.07, "learning_rate": 0.0004715346038114343, "loss": 3.3079, "theoretical_loss": 4.306427832849394, "tokens_seen": 219021312 }, { "epoch": 0.07, "learning_rate": 0.0004715245737211635, "loss": 3.19, "theoretical_loss": 4.306268884161959, "tokens_seen": 219086848 }, { "epoch": 0.07, "learning_rate": 0.0004715145436308927, "loss": 3.3318, "theoretical_loss": 4.306109996322679, "tokens_seen": 219152384 }, { "epoch": 0.07, "learning_rate": 0.00047150451354062184, "loss": 3.0093, "theoretical_loss": 4.305951169290073, "tokens_seen": 219217920 }, { "epoch": 0.07, "learning_rate": 0.0004714944834503511, "loss": 3.3021, "theoretical_loss": 4.305792403022703, "tokens_seen": 219283456 }, { "epoch": 0.07, "learning_rate": 0.0004714844533600802, "loss": 3.0608, "theoretical_loss": 4.305633697479171, "tokens_seen": 219348992 }, { "epoch": 0.07, "learning_rate": 0.00047147442326980944, "loss": 2.8014, "theoretical_loss": 4.305475052618119, "tokens_seen": 219414528 }, { "epoch": 0.07, "learning_rate": 0.0004714643931795387, "loss": 3.3019, "theoretical_loss": 4.30531646839823, "tokens_seen": 219480064 }, { "epoch": 0.07, "objective/train/docs_used": 135676, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.066269874572754, "objective/train/theoretical_loss": 4.305157944778228, "objective/train/tokens_used": 240005600, "theoretical_loss": 4.305157944778228, "tokens_seen": 219545600 }, { "epoch": 0.07, "learning_rate": 0.0004714543630892678, "loss": 3.0659, "theoretical_loss": 4.305157944778228, "tokens_seen": 219545600 }, { "epoch": 0.07, "learning_rate": 0.00047144433299899704, "loss": 3.251, "theoretical_loss": 4.304999481716876, "tokens_seen": 219611136 }, { "epoch": 0.07, "learning_rate": 0.00047143430290872616, "loss": 3.4636, "theoretical_loss": 4.304841079172979, "tokens_seen": 219676672 }, { "epoch": 0.07, "learning_rate": 0.0004714242728184554, "loss": 3.2243, "theoretical_loss": 4.30468273710538, "tokens_seen": 219742208 }, { "epoch": 0.07, "learning_rate": 0.0004714142427281846, "loss": 3.2012, "theoretical_loss": 4.304524455472965, "tokens_seen": 219807744 }, { "epoch": 0.07, "learning_rate": 0.00047140421263791376, "loss": 2.9528, "theoretical_loss": 4.304366234234659, "tokens_seen": 219873280 }, { "epoch": 0.07, "learning_rate": 0.00047139418254764294, "loss": 3.214, "theoretical_loss": 4.304208073349426, "tokens_seen": 219938816 }, { "epoch": 0.07, "learning_rate": 0.0004713841524573721, "loss": 3.3243, "theoretical_loss": 4.304049972776271, "tokens_seen": 220004352 }, { "epoch": 0.07, "learning_rate": 0.0004713741223671013, "loss": 3.169, "theoretical_loss": 4.30389193247424, "tokens_seen": 220069888 }, { "epoch": 0.07, "learning_rate": 0.00047136409227683054, "loss": 3.2084, "theoretical_loss": 4.303733952402419, "tokens_seen": 220135424 }, { "epoch": 0.07, "learning_rate": 0.00047135406218655967, "loss": 3.2908, "theoretical_loss": 4.303576032519931, "tokens_seen": 220200960 }, { "epoch": 0.07, "learning_rate": 0.0004713440320962889, "loss": 3.415, "theoretical_loss": 4.303418172785943, "tokens_seen": 220266496 }, { "epoch": 0.07, "learning_rate": 0.0004713340020060181, "loss": 3.3773, "theoretical_loss": 4.303260373159659, "tokens_seen": 220332032 }, { "epoch": 0.07, "learning_rate": 0.00047132397191574726, "loss": 3.2734, "theoretical_loss": 4.303102633600322, "tokens_seen": 220397568 }, { "epoch": 0.07, "learning_rate": 0.00047131394182547645, "loss": 2.8756, "theoretical_loss": 4.30294495406722, "tokens_seen": 220463104 }, { "epoch": 0.07, "learning_rate": 0.00047130391173520563, "loss": 3.1967, "theoretical_loss": 4.3027873345196745, "tokens_seen": 220528640 }, { "epoch": 0.07, "learning_rate": 0.0004712938816449348, "loss": 3.4082, "theoretical_loss": 4.302629774917049, "tokens_seen": 220594176 }, { "epoch": 0.07, "learning_rate": 0.00047128385155466404, "loss": 3.1634, "theoretical_loss": 4.302472275218748, "tokens_seen": 220659712 }, { "epoch": 0.07, "learning_rate": 0.00047127382146439317, "loss": 3.3901, "theoretical_loss": 4.302314835384214, "tokens_seen": 220725248 }, { "epoch": 0.07, "learning_rate": 0.0004712637913741224, "loss": 3.3387, "theoretical_loss": 4.30215745537293, "tokens_seen": 220790784 }, { "epoch": 0.07, "learning_rate": 0.00047125376128385153, "loss": 3.2651, "theoretical_loss": 4.302000135144416, "tokens_seen": 220856320 }, { "epoch": 0.07, "learning_rate": 0.00047124373119358077, "loss": 3.2082, "theoretical_loss": 4.301842874658235, "tokens_seen": 220921856 }, { "epoch": 0.07, "learning_rate": 0.00047123370110330995, "loss": 3.2934, "theoretical_loss": 4.301685673873987, "tokens_seen": 220987392 }, { "epoch": 0.07, "learning_rate": 0.00047122367101303913, "loss": 3.1771, "theoretical_loss": 4.301528532751312, "tokens_seen": 221052928 }, { "epoch": 0.07, "learning_rate": 0.0004712136409227683, "loss": 3.3551, "theoretical_loss": 4.301371451249888, "tokens_seen": 221118464 }, { "epoch": 0.07, "objective/train/docs_used": 136371, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.877000093460083, "objective/train/theoretical_loss": 4.301214429329433, "objective/train/tokens_used": 241644000, "theoretical_loss": 4.301214429329433, "tokens_seen": 221184000 }, { "epoch": 0.07, "learning_rate": 0.00047120361083249755, "loss": 3.4952, "theoretical_loss": 4.301214429329433, "tokens_seen": 221184000 }, { "epoch": 0.07, "learning_rate": 0.0004711935807422267, "loss": 3.1406, "theoretical_loss": 4.301057466949707, "tokens_seen": 221249536 }, { "epoch": 0.07, "learning_rate": 0.0004711835506519559, "loss": 3.1892, "theoretical_loss": 4.300900564070504, "tokens_seen": 221315072 }, { "epoch": 0.07, "learning_rate": 0.00047117352056168504, "loss": 2.9775, "theoretical_loss": 4.30074372065166, "tokens_seen": 221380608 }, { "epoch": 0.07, "learning_rate": 0.00047116349047141427, "loss": 3.3187, "theoretical_loss": 4.300586936653049, "tokens_seen": 221446144 }, { "epoch": 0.07, "learning_rate": 0.00047115346038114345, "loss": 3.148, "theoretical_loss": 4.300430212034587, "tokens_seen": 221511680 }, { "epoch": 0.07, "learning_rate": 0.00047114343029087263, "loss": 2.8728, "theoretical_loss": 4.300273546756223, "tokens_seen": 221577216 }, { "epoch": 0.07, "learning_rate": 0.0004711334002006018, "loss": 3.1421, "theoretical_loss": 4.300116940777951, "tokens_seen": 221642752 }, { "epoch": 0.07, "learning_rate": 0.000471123370110331, "loss": 3.1462, "theoretical_loss": 4.299960394059799, "tokens_seen": 221708288 }, { "epoch": 0.07, "learning_rate": 0.0004711133400200602, "loss": 2.9714, "theoretical_loss": 4.299803906561835, "tokens_seen": 221773824 }, { "epoch": 0.07, "learning_rate": 0.0004711033099297894, "loss": 3.278, "theoretical_loss": 4.29964747824417, "tokens_seen": 221839360 }, { "epoch": 0.07, "learning_rate": 0.00047109327983951854, "loss": 3.1696, "theoretical_loss": 4.299491109066947, "tokens_seen": 221904896 }, { "epoch": 0.07, "learning_rate": 0.0004710832497492478, "loss": 3.1112, "theoretical_loss": 4.299334798990351, "tokens_seen": 221970432 }, { "epoch": 0.07, "learning_rate": 0.0004710732196589769, "loss": 3.1501, "theoretical_loss": 4.2991785479746065, "tokens_seen": 222035968 }, { "epoch": 0.07, "learning_rate": 0.00047106318956870614, "loss": 2.7828, "theoretical_loss": 4.299022355979974, "tokens_seen": 222101504 }, { "epoch": 0.07, "learning_rate": 0.0004710531594784353, "loss": 3.3247, "theoretical_loss": 4.298866222966755, "tokens_seen": 222167040 }, { "epoch": 0.07, "learning_rate": 0.0004710431293881645, "loss": 3.1744, "theoretical_loss": 4.298710148895286, "tokens_seen": 222232576 }, { "epoch": 0.07, "learning_rate": 0.0004710330992978937, "loss": 3.4915, "theoretical_loss": 4.298554133725946, "tokens_seen": 222298112 }, { "epoch": 0.07, "learning_rate": 0.0004710230692076229, "loss": 3.1641, "theoretical_loss": 4.298398177419149, "tokens_seen": 222363648 }, { "epoch": 0.07, "learning_rate": 0.00047101303911735204, "loss": 3.2386, "theoretical_loss": 4.298242279935349, "tokens_seen": 222429184 }, { "epoch": 0.07, "learning_rate": 0.0004710030090270813, "loss": 3.305, "theoretical_loss": 4.2980864412350375, "tokens_seen": 222494720 }, { "epoch": 0.07, "learning_rate": 0.0004709929789368104, "loss": 3.1451, "theoretical_loss": 4.297930661278745, "tokens_seen": 222560256 }, { "epoch": 0.07, "learning_rate": 0.00047098294884653964, "loss": 3.031, "theoretical_loss": 4.297774940027038, "tokens_seen": 222625792 }, { "epoch": 0.07, "learning_rate": 0.0004709729187562688, "loss": 3.2461, "theoretical_loss": 4.297619277440523, "tokens_seen": 222691328 }, { "epoch": 0.07, "learning_rate": 0.000470962888665998, "loss": 3.1183, "theoretical_loss": 4.297463673479846, "tokens_seen": 222756864 }, { "epoch": 0.07, "objective/train/docs_used": 137650, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3066697120666504, "objective/train/theoretical_loss": 4.297308128105687, "objective/train/tokens_used": 243282400, "theoretical_loss": 4.297308128105687, "tokens_seen": 222822400 }, { "epoch": 0.07, "learning_rate": 0.0004709528585757272, "loss": 3.3301, "theoretical_loss": 4.297308128105687, "tokens_seen": 222822400 }, { "epoch": 0.07, "learning_rate": 0.00047094282848545636, "loss": 3.2327, "theoretical_loss": 4.297152641278767, "tokens_seen": 222887936 }, { "epoch": 0.07, "learning_rate": 0.00047093279839518555, "loss": 3.0477, "theoretical_loss": 4.296997212959842, "tokens_seen": 222953472 }, { "epoch": 0.07, "learning_rate": 0.0004709227683049148, "loss": 3.4307, "theoretical_loss": 4.296841843109711, "tokens_seen": 223019008 }, { "epoch": 0.07, "learning_rate": 0.0004709127382146439, "loss": 3.1878, "theoretical_loss": 4.296686531689204, "tokens_seen": 223084544 }, { "epoch": 0.07, "learning_rate": 0.00047090270812437314, "loss": 3.29, "theoretical_loss": 4.296531278659193, "tokens_seen": 223150080 }, { "epoch": 0.07, "learning_rate": 0.00047089267803410227, "loss": 3.0173, "theoretical_loss": 4.296376083980589, "tokens_seen": 223215616 }, { "epoch": 0.07, "learning_rate": 0.0004708826479438315, "loss": 3.3376, "theoretical_loss": 4.296220947614337, "tokens_seen": 223281152 }, { "epoch": 0.07, "learning_rate": 0.0004708726178535607, "loss": 2.9857, "theoretical_loss": 4.296065869521421, "tokens_seen": 223346688 }, { "epoch": 0.07, "learning_rate": 0.00047086258776328987, "loss": 3.5665, "theoretical_loss": 4.295910849662862, "tokens_seen": 223412224 }, { "epoch": 0.07, "learning_rate": 0.00047085255767301905, "loss": 3.2103, "theoretical_loss": 4.2957558879997215, "tokens_seen": 223477760 }, { "epoch": 0.07, "learning_rate": 0.0004708425275827483, "loss": 3.1934, "theoretical_loss": 4.295600984493093, "tokens_seen": 223543296 }, { "epoch": 0.07, "learning_rate": 0.0004708324974924774, "loss": 2.9751, "theoretical_loss": 4.295446139104112, "tokens_seen": 223608832 }, { "epoch": 0.07, "learning_rate": 0.00047082246740220665, "loss": 3.18, "theoretical_loss": 4.295291351793951, "tokens_seen": 223674368 }, { "epoch": 0.07, "learning_rate": 0.0004708124373119358, "loss": 3.1672, "theoretical_loss": 4.295136622523817, "tokens_seen": 223739904 }, { "epoch": 0.07, "learning_rate": 0.000470802407221665, "loss": 3.001, "theoretical_loss": 4.294981951254956, "tokens_seen": 223805440 }, { "epoch": 0.07, "learning_rate": 0.0004707923771313942, "loss": 3.1423, "theoretical_loss": 4.294827337948651, "tokens_seen": 223870976 }, { "epoch": 0.07, "learning_rate": 0.00047078234704112337, "loss": 3.0418, "theoretical_loss": 4.294672782566224, "tokens_seen": 223936512 }, { "epoch": 0.07, "learning_rate": 0.00047077231695085255, "loss": 3.1785, "theoretical_loss": 4.29451828506903, "tokens_seen": 224002048 }, { "epoch": 0.07, "learning_rate": 0.00047076228686058173, "loss": 3.2824, "theoretical_loss": 4.294363845418465, "tokens_seen": 224067584 }, { "epoch": 0.07, "learning_rate": 0.0004707522567703109, "loss": 3.099, "theoretical_loss": 4.29420946357596, "tokens_seen": 224133120 }, { "epoch": 0.07, "learning_rate": 0.00047074222668004015, "loss": 3.1006, "theoretical_loss": 4.294055139502985, "tokens_seen": 224198656 }, { "epoch": 0.07, "learning_rate": 0.0004707321965897693, "loss": 3.2409, "theoretical_loss": 4.293900873161043, "tokens_seen": 224264192 }, { "epoch": 0.07, "learning_rate": 0.0004707221664994985, "loss": 3.0344, "theoretical_loss": 4.293746664511678, "tokens_seen": 224329728 }, { "epoch": 0.07, "learning_rate": 0.00047071213640922775, "loss": 2.9979, "theoretical_loss": 4.293592513516469, "tokens_seen": 224395264 }, { "epoch": 0.07, "objective/train/docs_used": 138458, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.281918525695801, "objective/train/theoretical_loss": 4.293438420137031, "objective/train/tokens_used": 244920800, "theoretical_loss": 4.293438420137031, "tokens_seen": 224460800 }, { "epoch": 0.07, "learning_rate": 0.0004707021063189569, "loss": 3.1837, "theoretical_loss": 4.293438420137031, "tokens_seen": 224460800 }, { "epoch": 0.07, "learning_rate": 0.0004706920762286861, "loss": 3.0455, "theoretical_loss": 4.293284384335017, "tokens_seen": 224526336 }, { "epoch": 0.07, "learning_rate": 0.00047068204613841524, "loss": 3.2542, "theoretical_loss": 4.293130406072118, "tokens_seen": 224591872 }, { "epoch": 0.07, "learning_rate": 0.00047067201604814447, "loss": 3.2776, "theoretical_loss": 4.292976485310057, "tokens_seen": 224657408 }, { "epoch": 0.07, "learning_rate": 0.00047066198595787365, "loss": 3.1208, "theoretical_loss": 4.2928226220106005, "tokens_seen": 224722944 }, { "epoch": 0.07, "learning_rate": 0.00047065195586760283, "loss": 3.1642, "theoretical_loss": 4.292668816135545, "tokens_seen": 224788480 }, { "epoch": 0.07, "learning_rate": 0.000470641925777332, "loss": 3.3477, "theoretical_loss": 4.292515067646727, "tokens_seen": 224854016 }, { "epoch": 0.07, "learning_rate": 0.0004706318956870612, "loss": 3.1525, "theoretical_loss": 4.29236137650602, "tokens_seen": 224919552 }, { "epoch": 0.07, "learning_rate": 0.0004706218655967904, "loss": 3.203, "theoretical_loss": 4.2922077426753305, "tokens_seen": 224985088 }, { "epoch": 0.07, "learning_rate": 0.0004706118355065196, "loss": 3.304, "theoretical_loss": 4.292054166116605, "tokens_seen": 225050624 }, { "epoch": 0.07, "learning_rate": 0.00047060180541624874, "loss": 3.099, "theoretical_loss": 4.291900646791825, "tokens_seen": 225116160 }, { "epoch": 0.07, "learning_rate": 0.000470591775325978, "loss": 3.4831, "theoretical_loss": 4.2917471846630075, "tokens_seen": 225181696 }, { "epoch": 0.07, "learning_rate": 0.0004705817452357071, "loss": 3.3273, "theoretical_loss": 4.291593779692207, "tokens_seen": 225247232 }, { "epoch": 0.07, "learning_rate": 0.00047057171514543634, "loss": 3.0792, "theoretical_loss": 4.291440431841513, "tokens_seen": 225312768 }, { "epoch": 0.07, "learning_rate": 0.0004705616850551655, "loss": 3.1787, "theoretical_loss": 4.291287141073053, "tokens_seen": 225378304 }, { "epoch": 0.07, "learning_rate": 0.0004705516549648947, "loss": 3.0933, "theoretical_loss": 4.291133907348989, "tokens_seen": 225443840 }, { "epoch": 0.07, "learning_rate": 0.0004705416248746239, "loss": 3.2376, "theoretical_loss": 4.29098073063152, "tokens_seen": 225509376 }, { "epoch": 0.07, "learning_rate": 0.0004705315947843531, "loss": 3.3523, "theoretical_loss": 4.29082761088288, "tokens_seen": 225574912 }, { "epoch": 0.07, "learning_rate": 0.00047052156469408224, "loss": 3.032, "theoretical_loss": 4.290674548065338, "tokens_seen": 225640448 }, { "epoch": 0.07, "learning_rate": 0.0004705115346038115, "loss": 3.0997, "theoretical_loss": 4.290521542141203, "tokens_seen": 225705984 }, { "epoch": 0.07, "learning_rate": 0.0004705015045135406, "loss": 3.1698, "theoretical_loss": 4.290368593072817, "tokens_seen": 225771520 }, { "epoch": 0.07, "learning_rate": 0.00047049147442326984, "loss": 3.2039, "theoretical_loss": 4.290215700822556, "tokens_seen": 225837056 }, { "epoch": 0.07, "learning_rate": 0.000470481444332999, "loss": 3.1309, "theoretical_loss": 4.290062865352837, "tokens_seen": 225902592 }, { "epoch": 0.07, "learning_rate": 0.0004704714142427282, "loss": 3.1847, "theoretical_loss": 4.289910086626108, "tokens_seen": 225968128 }, { "epoch": 0.07, "learning_rate": 0.0004704613841524574, "loss": 2.9217, "theoretical_loss": 4.289757364604855, "tokens_seen": 226033664 }, { "epoch": 0.07, "objective/train/docs_used": 139641, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6932528018951416, "objective/train/theoretical_loss": 4.2896046992515995, "objective/train/tokens_used": 246559200, "theoretical_loss": 4.2896046992515995, "tokens_seen": 226099200 }, { "epoch": 0.07, "learning_rate": 0.00047045135406218656, "loss": 3.5236, "theoretical_loss": 4.2896046992515995, "tokens_seen": 226099200 }, { "epoch": 0.07, "learning_rate": 0.00047044132397191575, "loss": 3.1005, "theoretical_loss": 4.289452090528897, "tokens_seen": 226164736 }, { "epoch": 0.07, "learning_rate": 0.000470431293881645, "loss": 3.1343, "theoretical_loss": 4.289299538399341, "tokens_seen": 226230272 }, { "epoch": 0.07, "learning_rate": 0.0004704212637913741, "loss": 3.3457, "theoretical_loss": 4.28914704282556, "tokens_seen": 226295808 }, { "epoch": 0.07, "learning_rate": 0.00047041123370110334, "loss": 3.4157, "theoretical_loss": 4.288994603770215, "tokens_seen": 226361344 }, { "epoch": 0.07, "learning_rate": 0.00047040120361083247, "loss": 3.2237, "theoretical_loss": 4.288842221196007, "tokens_seen": 226426880 }, { "epoch": 0.07, "learning_rate": 0.0004703911735205617, "loss": 3.2044, "theoretical_loss": 4.28868989506567, "tokens_seen": 226492416 }, { "epoch": 0.07, "learning_rate": 0.0004703811434302909, "loss": 3.2856, "theoretical_loss": 4.288537625341974, "tokens_seen": 226557952 }, { "epoch": 0.07, "learning_rate": 0.00047037111334002007, "loss": 3.0679, "theoretical_loss": 4.288385411987722, "tokens_seen": 226623488 }, { "epoch": 0.07, "learning_rate": 0.00047036108324974925, "loss": 3.4104, "theoretical_loss": 4.288233254965755, "tokens_seen": 226689024 }, { "epoch": 0.07, "learning_rate": 0.0004703510531594785, "loss": 3.3151, "theoretical_loss": 4.2880811542389505, "tokens_seen": 226754560 }, { "epoch": 0.07, "learning_rate": 0.0004703410230692076, "loss": 3.3593, "theoretical_loss": 4.287929109770217, "tokens_seen": 226820096 }, { "epoch": 0.07, "learning_rate": 0.00047033099297893685, "loss": 2.8766, "theoretical_loss": 4.287777121522501, "tokens_seen": 226885632 }, { "epoch": 0.07, "learning_rate": 0.000470320962888666, "loss": 3.209, "theoretical_loss": 4.287625189458781, "tokens_seen": 226951168 }, { "epoch": 0.07, "learning_rate": 0.0004703109327983952, "loss": 3.1, "theoretical_loss": 4.287473313542077, "tokens_seen": 227016704 }, { "epoch": 0.07, "learning_rate": 0.0004703009027081244, "loss": 3.3124, "theoretical_loss": 4.287321493735438, "tokens_seen": 227082240 }, { "epoch": 0.07, "learning_rate": 0.00047029087261785357, "loss": 3.1342, "theoretical_loss": 4.287169730001949, "tokens_seen": 227147776 }, { "epoch": 0.07, "learning_rate": 0.00047028084252758275, "loss": 3.1754, "theoretical_loss": 4.287018022304733, "tokens_seen": 227213312 }, { "epoch": 0.07, "learning_rate": 0.00047027081243731193, "loss": 3.2797, "theoretical_loss": 4.286866370606943, "tokens_seen": 227278848 }, { "epoch": 0.07, "learning_rate": 0.0004702607823470411, "loss": 3.3938, "theoretical_loss": 4.286714774871772, "tokens_seen": 227344384 }, { "epoch": 0.07, "learning_rate": 0.00047025075225677035, "loss": 2.8743, "theoretical_loss": 4.286563235062444, "tokens_seen": 227409920 }, { "epoch": 0.07, "learning_rate": 0.0004702407221664995, "loss": 3.2675, "theoretical_loss": 4.28641175114222, "tokens_seen": 227475456 }, { "epoch": 0.07, "learning_rate": 0.0004702306920762287, "loss": 3.5072, "theoretical_loss": 4.286260323074394, "tokens_seen": 227540992 }, { "epoch": 0.07, "learning_rate": 0.00047022066198595784, "loss": 3.385, "theoretical_loss": 4.286108950822296, "tokens_seen": 227606528 }, { "epoch": 0.07, "learning_rate": 0.0004702106318956871, "loss": 3.2283, "theoretical_loss": 4.285957634349289, "tokens_seen": 227672064 }, { "epoch": 0.07, "objective/train/docs_used": 140376, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.265362501144409, "objective/train/theoretical_loss": 4.285806373618774, "objective/train/tokens_used": 248197600, "theoretical_loss": 4.285806373618774, "tokens_seen": 227737600 }, { "epoch": 0.07, "learning_rate": 0.00047020060180541626, "loss": 2.999, "theoretical_loss": 4.285806373618774, "tokens_seen": 227737600 }, { "epoch": 0.07, "learning_rate": 0.00047019057171514544, "loss": 3.1608, "theoretical_loss": 4.285655168594182, "tokens_seen": 227803136 }, { "epoch": 0.07, "learning_rate": 0.0004701805416248746, "loss": 3.2796, "theoretical_loss": 4.285504019238982, "tokens_seen": 227868672 }, { "epoch": 0.07, "learning_rate": 0.00047017051153460385, "loss": 3.4479, "theoretical_loss": 4.285352925516676, "tokens_seen": 227934208 }, { "epoch": 0.07, "learning_rate": 0.000470160481444333, "loss": 3.3548, "theoretical_loss": 4.2852018873908, "tokens_seen": 227999744 }, { "epoch": 0.07, "learning_rate": 0.0004701504513540622, "loss": 3.2998, "theoretical_loss": 4.285050904824925, "tokens_seen": 228065280 }, { "epoch": 0.07, "learning_rate": 0.00047014042126379134, "loss": 3.1287, "theoretical_loss": 4.284899977782658, "tokens_seen": 228130816 }, { "epoch": 0.07, "learning_rate": 0.0004701303911735206, "loss": 3.2708, "theoretical_loss": 4.284749106227636, "tokens_seen": 228196352 }, { "epoch": 0.07, "learning_rate": 0.00047012036108324976, "loss": 3.1895, "theoretical_loss": 4.284598290123535, "tokens_seen": 228261888 }, { "epoch": 0.07, "learning_rate": 0.00047011033099297894, "loss": 3.128, "theoretical_loss": 4.284447529434061, "tokens_seen": 228327424 }, { "epoch": 0.07, "learning_rate": 0.0004701003009027081, "loss": 3.2905, "theoretical_loss": 4.284296824122959, "tokens_seen": 228392960 }, { "epoch": 0.07, "learning_rate": 0.0004700902708124373, "loss": 3.152, "theoretical_loss": 4.284146174154003, "tokens_seen": 228458496 }, { "epoch": 0.07, "learning_rate": 0.0004700802407221665, "loss": 3.4164, "theoretical_loss": 4.283995579491004, "tokens_seen": 228524032 }, { "epoch": 0.07, "learning_rate": 0.0004700702106318957, "loss": 2.9542, "theoretical_loss": 4.283845040097807, "tokens_seen": 228589568 }, { "epoch": 0.07, "learning_rate": 0.00047006018054162484, "loss": 3.1201, "theoretical_loss": 4.28369455593829, "tokens_seen": 228655104 }, { "epoch": 0.07, "learning_rate": 0.0004700501504513541, "loss": 3.3299, "theoretical_loss": 4.2835441269763646, "tokens_seen": 228720640 }, { "epoch": 0.07, "learning_rate": 0.00047004012036108326, "loss": 2.9468, "theoretical_loss": 4.283393753175979, "tokens_seen": 228786176 }, { "epoch": 0.07, "learning_rate": 0.00047003009027081244, "loss": 3.3356, "theoretical_loss": 4.283243434501112, "tokens_seen": 228851712 }, { "epoch": 0.07, "learning_rate": 0.0004700200601805416, "loss": 3.0321, "theoretical_loss": 4.283093170915778, "tokens_seen": 228917248 }, { "epoch": 0.07, "learning_rate": 0.0004700100300902708, "loss": 3.143, "theoretical_loss": 4.282942962384023, "tokens_seen": 228982784 }, { "epoch": 0.07, "learning_rate": 0.00047, "loss": 3.1754, "theoretical_loss": 4.282792808869932, "tokens_seen": 229048320 }, { "epoch": 0.07, "learning_rate": 0.0004699899699097292, "loss": 3.1478, "theoretical_loss": 4.282642710337618, "tokens_seen": 229113856 }, { "epoch": 0.07, "learning_rate": 0.00046997993981945835, "loss": 3.336, "theoretical_loss": 4.28249266675123, "tokens_seen": 229179392 }, { "epoch": 0.07, "learning_rate": 0.0004699699097291876, "loss": 3.2775, "theoretical_loss": 4.282342678074951, "tokens_seen": 229244928 }, { "epoch": 0.07, "learning_rate": 0.00046995987963891676, "loss": 3.2018, "theoretical_loss": 4.2821927442729955, "tokens_seen": 229310464 }, { "epoch": 0.07, "objective/train/docs_used": 141693, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1491169929504395, "objective/train/theoretical_loss": 4.282042865309616, "objective/train/tokens_used": 249836000, "theoretical_loss": 4.282042865309616, "tokens_seen": 229376000 }, { "epoch": 0.07, "learning_rate": 0.00046994984954864595, "loss": 3.1859, "theoretical_loss": 4.282042865309616, "tokens_seen": 229376000 }, { "epoch": 0.07, "learning_rate": 0.0004699398194583752, "loss": 3.3997, "theoretical_loss": 4.281893041149093, "tokens_seen": 229441536 }, { "epoch": 0.07, "learning_rate": 0.0004699297893681043, "loss": 3.1647, "theoretical_loss": 4.2817432717557455, "tokens_seen": 229507072 }, { "epoch": 0.07, "learning_rate": 0.00046991975927783354, "loss": 3.2627, "theoretical_loss": 4.28159355709392, "tokens_seen": 229572608 }, { "epoch": 0.07, "learning_rate": 0.00046990972918756267, "loss": 3.1173, "theoretical_loss": 4.281443897128004, "tokens_seen": 229638144 }, { "epoch": 0.07, "learning_rate": 0.0004698996990972919, "loss": 3.3119, "theoretical_loss": 4.2812942918224115, "tokens_seen": 229703680 }, { "epoch": 0.07, "learning_rate": 0.0004698896690070211, "loss": 2.8788, "theoretical_loss": 4.281144741141593, "tokens_seen": 229769216 }, { "epoch": 0.07, "learning_rate": 0.00046987963891675027, "loss": 3.2099, "theoretical_loss": 4.280995245050032, "tokens_seen": 229834752 }, { "epoch": 0.07, "learning_rate": 0.00046986960882647945, "loss": 3.1133, "theoretical_loss": 4.2808458035122445, "tokens_seen": 229900288 }, { "epoch": 0.07, "learning_rate": 0.0004698595787362087, "loss": 3.3331, "theoretical_loss": 4.2806964164927805, "tokens_seen": 229965824 }, { "epoch": 0.07, "learning_rate": 0.0004698495486459378, "loss": 3.0517, "theoretical_loss": 4.280547083956224, "tokens_seen": 230031360 }, { "epoch": 0.07, "learning_rate": 0.00046983951855566705, "loss": 3.1513, "theoretical_loss": 4.280397805867188, "tokens_seen": 230096896 }, { "epoch": 0.07, "learning_rate": 0.0004698294884653962, "loss": 3.1371, "theoretical_loss": 4.280248582190324, "tokens_seen": 230162432 }, { "epoch": 0.07, "learning_rate": 0.0004698194583751254, "loss": 3.1632, "theoretical_loss": 4.280099412890312, "tokens_seen": 230227968 }, { "epoch": 0.07, "learning_rate": 0.0004698094282848546, "loss": 3.1009, "theoretical_loss": 4.279950297931869, "tokens_seen": 230293504 }, { "epoch": 0.07, "learning_rate": 0.00046979939819458377, "loss": 3.414, "theoretical_loss": 4.27980123727974, "tokens_seen": 230359040 }, { "epoch": 0.07, "learning_rate": 0.00046978936810431295, "loss": 3.1625, "theoretical_loss": 4.279652230898709, "tokens_seen": 230424576 }, { "epoch": 0.07, "learning_rate": 0.00046977933801404213, "loss": 3.2313, "theoretical_loss": 4.279503278753586, "tokens_seen": 230490112 }, { "epoch": 0.07, "learning_rate": 0.0004697693079237713, "loss": 3.3858, "theoretical_loss": 4.27935438080922, "tokens_seen": 230555648 }, { "epoch": 0.07, "learning_rate": 0.00046975927783350055, "loss": 3.011, "theoretical_loss": 4.27920553703049, "tokens_seen": 230621184 }, { "epoch": 0.07, "learning_rate": 0.0004697492477432297, "loss": 2.9844, "theoretical_loss": 4.279056747382306, "tokens_seen": 230686720 }, { "epoch": 0.07, "learning_rate": 0.0004697392176529589, "loss": 3.197, "theoretical_loss": 4.278908011829613, "tokens_seen": 230752256 }, { "epoch": 0.07, "learning_rate": 0.00046972918756268804, "loss": 3.1954, "theoretical_loss": 4.27875933033739, "tokens_seen": 230817792 }, { "epoch": 0.07, "learning_rate": 0.0004697191574724173, "loss": 3.0489, "theoretical_loss": 4.278610702870646, "tokens_seen": 230883328 }, { "epoch": 0.07, "learning_rate": 0.00046970912738214646, "loss": 3.2719, "theoretical_loss": 4.278462129394423, "tokens_seen": 230948864 }, { "epoch": 0.07, "objective/train/docs_used": 142407, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.702136993408203, "objective/train/theoretical_loss": 4.278313609873795, "objective/train/tokens_used": 251474400, "theoretical_loss": 4.278313609873795, "tokens_seen": 231014400 }, { "epoch": 0.07, "learning_rate": 0.00046969909729187564, "loss": 3.3355, "theoretical_loss": 4.278313609873795, "tokens_seen": 231014400 }, { "epoch": 0.07, "learning_rate": 0.0004696890672016048, "loss": 2.9092, "theoretical_loss": 4.278165144273871, "tokens_seen": 231079936 }, { "epoch": 0.07, "learning_rate": 0.00046967903711133405, "loss": 3.2064, "theoretical_loss": 4.27801673255979, "tokens_seen": 231145472 }, { "epoch": 0.07, "learning_rate": 0.0004696690070210632, "loss": 3.2198, "theoretical_loss": 4.277868374696725, "tokens_seen": 231211008 }, { "epoch": 0.07, "learning_rate": 0.0004696589769307924, "loss": 3.1205, "theoretical_loss": 4.277720070649879, "tokens_seen": 231276544 }, { "epoch": 0.07, "learning_rate": 0.00046964894684052154, "loss": 3.1119, "theoretical_loss": 4.277571820384491, "tokens_seen": 231342080 }, { "epoch": 0.07, "learning_rate": 0.0004696389167502508, "loss": 3.2283, "theoretical_loss": 4.277423623865829, "tokens_seen": 231407616 }, { "epoch": 0.07, "learning_rate": 0.00046962888665997996, "loss": 3.036, "theoretical_loss": 4.277275481059195, "tokens_seen": 231473152 }, { "epoch": 0.07, "learning_rate": 0.00046961885656970914, "loss": 3.0813, "theoretical_loss": 4.2771273919299215, "tokens_seen": 231538688 }, { "epoch": 0.07, "learning_rate": 0.0004696088264794383, "loss": 3.2151, "theoretical_loss": 4.276979356443377, "tokens_seen": 231604224 }, { "epoch": 0.07, "learning_rate": 0.0004695987963891675, "loss": 2.992, "theoretical_loss": 4.276831374564957, "tokens_seen": 231669760 }, { "epoch": 0.07, "learning_rate": 0.0004695887662988967, "loss": 3.1676, "theoretical_loss": 4.276683446260093, "tokens_seen": 231735296 }, { "epoch": 0.07, "learning_rate": 0.0004695787362086259, "loss": 3.0428, "theoretical_loss": 4.276535571494247, "tokens_seen": 231800832 }, { "epoch": 0.07, "learning_rate": 0.00046956870611835505, "loss": 2.979, "theoretical_loss": 4.276387750232913, "tokens_seen": 231866368 }, { "epoch": 0.07, "learning_rate": 0.0004695586760280843, "loss": 3.0738, "theoretical_loss": 4.276239982441617, "tokens_seen": 231931904 }, { "epoch": 0.07, "learning_rate": 0.00046954864593781346, "loss": 3.2279, "theoretical_loss": 4.276092268085918, "tokens_seen": 231997440 }, { "epoch": 0.07, "learning_rate": 0.00046953861584754264, "loss": 3.1162, "theoretical_loss": 4.275944607131406, "tokens_seen": 232062976 }, { "epoch": 0.07, "learning_rate": 0.0004695285857572718, "loss": 3.0761, "theoretical_loss": 4.275796999543703, "tokens_seen": 232128512 }, { "epoch": 0.07, "learning_rate": 0.000469518555667001, "loss": 3.0807, "theoretical_loss": 4.275649445288461, "tokens_seen": 232194048 }, { "epoch": 0.07, "learning_rate": 0.0004695085255767302, "loss": 3.1171, "theoretical_loss": 4.275501944331367, "tokens_seen": 232259584 }, { "epoch": 0.07, "learning_rate": 0.0004694984954864594, "loss": 3.273, "theoretical_loss": 4.275354496638139, "tokens_seen": 232325120 }, { "epoch": 0.07, "learning_rate": 0.00046948846539618855, "loss": 3.2311, "theoretical_loss": 4.275207102174525, "tokens_seen": 232390656 }, { "epoch": 0.07, "learning_rate": 0.0004694784353059178, "loss": 3.3267, "theoretical_loss": 4.275059760906305, "tokens_seen": 232456192 }, { "epoch": 0.07, "learning_rate": 0.0004694684052156469, "loss": 3.1457, "theoretical_loss": 4.2749124727992935, "tokens_seen": 232521728 }, { "epoch": 0.07, "learning_rate": 0.00046945837512537615, "loss": 3.4693, "theoretical_loss": 4.274765237819333, "tokens_seen": 232587264 }, { "epoch": 0.07, "objective/train/docs_used": 143093, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.266288995742798, "objective/train/theoretical_loss": 4.274618055932298, "objective/train/tokens_used": 253112800, "theoretical_loss": 4.274618055932298, "tokens_seen": 232652800 }, { "epoch": 0.07, "learning_rate": 0.0004694483450351053, "loss": 3.1205, "theoretical_loss": 4.274618055932298, "tokens_seen": 232652800 }, { "epoch": 0.07, "learning_rate": 0.0004694383149448345, "loss": 3.3176, "theoretical_loss": 4.2744709271040975, "tokens_seen": 232718336 }, { "epoch": 0.07, "learning_rate": 0.0004694282848545637, "loss": 3.1205, "theoretical_loss": 4.27432385130067, "tokens_seen": 232783872 }, { "epoch": 0.07, "learning_rate": 0.00046941825476429287, "loss": 3.1899, "theoretical_loss": 4.274176828487984, "tokens_seen": 232849408 }, { "epoch": 0.07, "learning_rate": 0.00046940822467402205, "loss": 3.2353, "theoretical_loss": 4.2740298586320415, "tokens_seen": 232914944 }, { "epoch": 0.07, "learning_rate": 0.0004693981945837513, "loss": 3.1096, "theoretical_loss": 4.273882941698876, "tokens_seen": 232980480 }, { "epoch": 0.07, "learning_rate": 0.0004693881644934804, "loss": 3.0017, "theoretical_loss": 4.27373607765455, "tokens_seen": 233046016 }, { "epoch": 0.07, "learning_rate": 0.00046937813440320965, "loss": 3.0144, "theoretical_loss": 4.2735892664651605, "tokens_seen": 233111552 }, { "epoch": 0.07, "learning_rate": 0.00046936810431293883, "loss": 3.2733, "theoretical_loss": 4.273442508096833, "tokens_seen": 233177088 }, { "epoch": 0.07, "learning_rate": 0.000469358074222668, "loss": 3.1154, "theoretical_loss": 4.273295802515726, "tokens_seen": 233242624 }, { "epoch": 0.07, "learning_rate": 0.0004693480441323972, "loss": 3.1081, "theoretical_loss": 4.273149149688028, "tokens_seen": 233308160 }, { "epoch": 0.07, "learning_rate": 0.0004693380140421264, "loss": 3.0391, "theoretical_loss": 4.27300254957996, "tokens_seen": 233373696 }, { "epoch": 0.07, "learning_rate": 0.00046932798395185555, "loss": 3.0994, "theoretical_loss": 4.272856002157772, "tokens_seen": 233439232 }, { "epoch": 0.07, "learning_rate": 0.0004693179538615848, "loss": 3.0702, "theoretical_loss": 4.272709507387748, "tokens_seen": 233504768 }, { "epoch": 0.07, "learning_rate": 0.0004693079237713139, "loss": 3.1512, "theoretical_loss": 4.2725630652362, "tokens_seen": 233570304 }, { "epoch": 0.07, "learning_rate": 0.00046929789368104315, "loss": 2.9167, "theoretical_loss": 4.272416675669473, "tokens_seen": 233635840 }, { "epoch": 0.07, "learning_rate": 0.0004692878635907723, "loss": 3.353, "theoretical_loss": 4.272270338653942, "tokens_seen": 233701376 }, { "epoch": 0.07, "learning_rate": 0.0004692778335005015, "loss": 3.0815, "theoretical_loss": 4.272124054156014, "tokens_seen": 233766912 }, { "epoch": 0.07, "learning_rate": 0.0004692678034102307, "loss": 3.1558, "theoretical_loss": 4.271977822142125, "tokens_seen": 233832448 }, { "epoch": 0.07, "learning_rate": 0.0004692577733199599, "loss": 3.4225, "theoretical_loss": 4.271831642578745, "tokens_seen": 233897984 }, { "epoch": 0.07, "learning_rate": 0.00046924774322968906, "loss": 3.0911, "theoretical_loss": 4.27168551543237, "tokens_seen": 233963520 }, { "epoch": 0.07, "learning_rate": 0.00046923771313941824, "loss": 3.1895, "theoretical_loss": 4.271539440669532, "tokens_seen": 234029056 }, { "epoch": 0.07, "learning_rate": 0.0004692276830491474, "loss": 3.223, "theoretical_loss": 4.27139341825679, "tokens_seen": 234094592 }, { "epoch": 0.07, "learning_rate": 0.00046921765295887666, "loss": 3.4205, "theoretical_loss": 4.271247448160736, "tokens_seen": 234160128 }, { "epoch": 0.07, "learning_rate": 0.00046920762286860584, "loss": 3.1894, "theoretical_loss": 4.27110153034799, "tokens_seen": 234225664 }, { "epoch": 0.07, "objective/train/docs_used": 144323, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.5741994380950928, "objective/train/theoretical_loss": 4.270955664785207, "objective/train/tokens_used": 254751200, "theoretical_loss": 4.270955664785207, "tokens_seen": 234291200 }, { "epoch": 0.07, "learning_rate": 0.000469197592778335, "loss": 3.2851, "theoretical_loss": 4.270955664785207, "tokens_seen": 234291200 }, { "epoch": 0.07, "learning_rate": 0.00046918756268806425, "loss": 3.207, "theoretical_loss": 4.2708098514390676, "tokens_seen": 234356736 }, { "epoch": 0.07, "learning_rate": 0.0004691775325977934, "loss": 3.0235, "theoretical_loss": 4.270664090276286, "tokens_seen": 234422272 }, { "epoch": 0.07, "learning_rate": 0.0004691675025075226, "loss": 3.2393, "theoretical_loss": 4.2705183812636065, "tokens_seen": 234487808 }, { "epoch": 0.07, "learning_rate": 0.00046915747241725174, "loss": 3.1636, "theoretical_loss": 4.270372724367803, "tokens_seen": 234553344 }, { "epoch": 0.07, "learning_rate": 0.000469147442326981, "loss": 3.073, "theoretical_loss": 4.270227119555681, "tokens_seen": 234618880 }, { "epoch": 0.07, "learning_rate": 0.00046913741223671016, "loss": 3.199, "theoretical_loss": 4.270081566794076, "tokens_seen": 234684416 }, { "epoch": 0.07, "learning_rate": 0.00046912738214643934, "loss": 3.0461, "theoretical_loss": 4.269936066049852, "tokens_seen": 234749952 }, { "epoch": 0.07, "learning_rate": 0.0004691173520561685, "loss": 3.0538, "theoretical_loss": 4.269790617289907, "tokens_seen": 234815488 }, { "epoch": 0.07, "learning_rate": 0.0004691073219658977, "loss": 3.0901, "theoretical_loss": 4.269645220481166, "tokens_seen": 234881024 }, { "epoch": 0.07, "learning_rate": 0.0004690972918756269, "loss": 3.1677, "theoretical_loss": 4.269499875590587, "tokens_seen": 234946560 }, { "epoch": 0.07, "learning_rate": 0.0004690872617853561, "loss": 3.1924, "theoretical_loss": 4.269354582585156, "tokens_seen": 235012096 }, { "epoch": 0.07, "learning_rate": 0.00046907723169508525, "loss": 2.9523, "theoretical_loss": 4.269209341431889, "tokens_seen": 235077632 }, { "epoch": 0.07, "learning_rate": 0.0004690672016048145, "loss": 2.9465, "theoretical_loss": 4.269064152097835, "tokens_seen": 235143168 }, { "epoch": 0.07, "learning_rate": 0.00046905717151454366, "loss": 3.2225, "theoretical_loss": 4.26891901455007, "tokens_seen": 235208704 }, { "epoch": 0.07, "learning_rate": 0.00046904714142427284, "loss": 3.3442, "theoretical_loss": 4.268773928755701, "tokens_seen": 235274240 }, { "epoch": 0.07, "learning_rate": 0.000469037111334002, "loss": 2.9913, "theoretical_loss": 4.268628894681868, "tokens_seen": 235339776 }, { "epoch": 0.07, "learning_rate": 0.0004690270812437312, "loss": 3.1332, "theoretical_loss": 4.268483912295735, "tokens_seen": 235405312 }, { "epoch": 0.07, "learning_rate": 0.0004690170511534604, "loss": 3.1891, "theoretical_loss": 4.268338981564502, "tokens_seen": 235470848 }, { "epoch": 0.07, "learning_rate": 0.0004690070210631896, "loss": 3.0551, "theoretical_loss": 4.268194102455395, "tokens_seen": 235536384 }, { "epoch": 0.07, "learning_rate": 0.00046899699097291875, "loss": 3.1104, "theoretical_loss": 4.26804927493567, "tokens_seen": 235601920 }, { "epoch": 0.07, "learning_rate": 0.000468986960882648, "loss": 3.1895, "theoretical_loss": 4.267904498972618, "tokens_seen": 235667456 }, { "epoch": 0.07, "learning_rate": 0.0004689769307923771, "loss": 3.115, "theoretical_loss": 4.267759774533552, "tokens_seen": 235732992 }, { "epoch": 0.07, "learning_rate": 0.00046896690070210635, "loss": 3.2606, "theoretical_loss": 4.267615101585821, "tokens_seen": 235798528 }, { "epoch": 0.07, "learning_rate": 0.00046895687061183553, "loss": 3.1975, "theoretical_loss": 4.267470480096801, "tokens_seen": 235864064 }, { "epoch": 0.07, "objective/train/docs_used": 145084, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.02962589263916, "objective/train/theoretical_loss": 4.267325910033897, "objective/train/tokens_used": 256389600, "theoretical_loss": 4.267325910033897, "tokens_seen": 235929600 }, { "epoch": 0.07, "learning_rate": 0.0004689468405215647, "loss": 3.2423, "theoretical_loss": 4.267325910033897, "tokens_seen": 235929600 }, { "epoch": 0.07, "learning_rate": 0.0004689368104312939, "loss": 3.0465, "theoretical_loss": 4.267181391364547, "tokens_seen": 235995136 }, { "epoch": 0.07, "learning_rate": 0.00046892678034102307, "loss": 3.0515, "theoretical_loss": 4.267036924056215, "tokens_seen": 236060672 }, { "epoch": 0.07, "learning_rate": 0.00046891675025075225, "loss": 3.1894, "theoretical_loss": 4.266892508076397, "tokens_seen": 236126208 }, { "epoch": 0.07, "learning_rate": 0.0004689067201604815, "loss": 3.3846, "theoretical_loss": 4.266748143392617, "tokens_seen": 236191744 }, { "epoch": 0.07, "learning_rate": 0.0004688966900702106, "loss": 2.8453, "theoretical_loss": 4.26660382997243, "tokens_seen": 236257280 }, { "epoch": 0.07, "learning_rate": 0.00046888665997993985, "loss": 3.2638, "theoretical_loss": 4.26645956778342, "tokens_seen": 236322816 }, { "epoch": 0.07, "learning_rate": 0.00046887662988966903, "loss": 3.2298, "theoretical_loss": 4.2663153567932, "tokens_seen": 236388352 }, { "epoch": 0.07, "learning_rate": 0.0004688665997993982, "loss": 3.4072, "theoretical_loss": 4.266171196969412, "tokens_seen": 236453888 }, { "epoch": 0.07, "learning_rate": 0.0004688565697091274, "loss": 3.3305, "theoretical_loss": 4.2660270882797295, "tokens_seen": 236519424 }, { "epoch": 0.07, "learning_rate": 0.0004688465396188566, "loss": 3.1997, "theoretical_loss": 4.265883030691853, "tokens_seen": 236584960 }, { "epoch": 0.07, "learning_rate": 0.00046883650952858575, "loss": 3.0699, "theoretical_loss": 4.265739024173515, "tokens_seen": 236650496 }, { "epoch": 0.07, "learning_rate": 0.000468826479438315, "loss": 3.1803, "theoretical_loss": 4.265595068692473, "tokens_seen": 236716032 }, { "epoch": 0.07, "learning_rate": 0.0004688164493480441, "loss": 3.254, "theoretical_loss": 4.26545116421652, "tokens_seen": 236781568 }, { "epoch": 0.07, "learning_rate": 0.00046880641925777335, "loss": 2.9222, "theoretical_loss": 4.265307310713471, "tokens_seen": 236847104 }, { "epoch": 0.07, "learning_rate": 0.0004687963891675025, "loss": 3.0569, "theoretical_loss": 4.2651635081511765, "tokens_seen": 236912640 }, { "epoch": 0.07, "learning_rate": 0.0004687863590772317, "loss": 3.1851, "theoretical_loss": 4.265019756497512, "tokens_seen": 236978176 }, { "epoch": 0.07, "learning_rate": 0.0004687763289869609, "loss": 3.0481, "theoretical_loss": 4.264876055720386, "tokens_seen": 237043712 }, { "epoch": 0.07, "learning_rate": 0.0004687662988966901, "loss": 3.2352, "theoretical_loss": 4.264732405787731, "tokens_seen": 237109248 }, { "epoch": 0.07, "learning_rate": 0.00046875626880641926, "loss": 2.9571, "theoretical_loss": 4.264588806667513, "tokens_seen": 237174784 }, { "epoch": 0.07, "learning_rate": 0.00046874623871614844, "loss": 3.07, "theoretical_loss": 4.264445258327724, "tokens_seen": 237240320 }, { "epoch": 0.07, "learning_rate": 0.0004687362086258776, "loss": 3.1563, "theoretical_loss": 4.264301760736389, "tokens_seen": 237305856 }, { "epoch": 0.07, "learning_rate": 0.00046872617853560686, "loss": 3.3152, "theoretical_loss": 4.264158313861557, "tokens_seen": 237371392 }, { "epoch": 0.07, "learning_rate": 0.000468716148445336, "loss": 3.1966, "theoretical_loss": 4.264014917671309, "tokens_seen": 237436928 }, { "epoch": 0.07, "learning_rate": 0.0004687061183550652, "loss": 3.0981, "theoretical_loss": 4.2638715721337554, "tokens_seen": 237502464 }, { "epoch": 0.07, "objective/train/docs_used": 146518, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.969031572341919, "objective/train/theoretical_loss": 4.263728277217032, "objective/train/tokens_used": 258028000, "theoretical_loss": 4.263728277217032, "tokens_seen": 237568000 }, { "epoch": 0.07, "learning_rate": 0.0004686960882647944, "loss": 2.995, "theoretical_loss": 4.263728277217032, "tokens_seen": 237568000 }, { "epoch": 0.07, "learning_rate": 0.0004686860581745236, "loss": 2.7096, "theoretical_loss": 4.263585032889306, "tokens_seen": 237633536 }, { "epoch": 0.07, "learning_rate": 0.00046867602808425276, "loss": 3.2779, "theoretical_loss": 4.263441839118776, "tokens_seen": 237699072 }, { "epoch": 0.07, "learning_rate": 0.00046866599799398194, "loss": 3.237, "theoretical_loss": 4.2632986958736625, "tokens_seen": 237764608 }, { "epoch": 0.07, "learning_rate": 0.0004686559679037111, "loss": 3.1594, "theoretical_loss": 4.263155603122221, "tokens_seen": 237830144 }, { "epoch": 0.07, "learning_rate": 0.00046864593781344036, "loss": 3.1277, "theoretical_loss": 4.263012560832733, "tokens_seen": 237895680 }, { "epoch": 0.07, "learning_rate": 0.0004686359077231695, "loss": 3.4282, "theoretical_loss": 4.262869568973508, "tokens_seen": 237961216 }, { "epoch": 0.07, "learning_rate": 0.0004686258776328987, "loss": 3.1725, "theoretical_loss": 4.262726627512886, "tokens_seen": 238026752 }, { "epoch": 0.07, "learning_rate": 0.00046861584754262785, "loss": 3.1387, "theoretical_loss": 4.262583736419234, "tokens_seen": 238092288 }, { "epoch": 0.07, "learning_rate": 0.0004686058174523571, "loss": 3.2668, "theoretical_loss": 4.26244089566095, "tokens_seen": 238157824 }, { "epoch": 0.07, "learning_rate": 0.00046859578736208626, "loss": 3.023, "theoretical_loss": 4.262298105206456, "tokens_seen": 238223360 }, { "epoch": 0.07, "learning_rate": 0.00046858575727181545, "loss": 3.4639, "theoretical_loss": 4.262155365024207, "tokens_seen": 238288896 }, { "epoch": 0.07, "learning_rate": 0.0004685757271815446, "loss": 3.1628, "theoretical_loss": 4.262012675082685, "tokens_seen": 238354432 }, { "epoch": 0.07, "learning_rate": 0.0004685656970912738, "loss": 3.0802, "theoretical_loss": 4.261870035350399, "tokens_seen": 238419968 }, { "epoch": 0.07, "learning_rate": 0.000468555667001003, "loss": 2.9962, "theoretical_loss": 4.261727445795888, "tokens_seen": 238485504 }, { "epoch": 0.07, "learning_rate": 0.0004685456369107322, "loss": 3.2431, "theoretical_loss": 4.26158490638772, "tokens_seen": 238551040 }, { "epoch": 0.07, "learning_rate": 0.00046853560682046135, "loss": 3.1662, "theoretical_loss": 4.261442417094488, "tokens_seen": 238616576 }, { "epoch": 0.07, "learning_rate": 0.0004685255767301906, "loss": 3.1555, "theoretical_loss": 4.261299977884816, "tokens_seen": 238682112 }, { "epoch": 0.07, "learning_rate": 0.00046851554663991977, "loss": 2.8705, "theoretical_loss": 4.2611575887273565, "tokens_seen": 238747648 }, { "epoch": 0.07, "learning_rate": 0.00046850551654964895, "loss": 2.9495, "theoretical_loss": 4.261015249590789, "tokens_seen": 238813184 }, { "epoch": 0.07, "learning_rate": 0.00046849548645937813, "loss": 3.0884, "theoretical_loss": 4.260872960443822, "tokens_seen": 238878720 }, { "epoch": 0.07, "learning_rate": 0.0004684854563691073, "loss": 3.192, "theoretical_loss": 4.260730721255191, "tokens_seen": 238944256 }, { "epoch": 0.07, "learning_rate": 0.0004684754262788365, "loss": 3.1981, "theoretical_loss": 4.260588531993662, "tokens_seen": 239009792 }, { "epoch": 0.07, "learning_rate": 0.00046846539618856573, "loss": 2.9505, "theoretical_loss": 4.260446392628026, "tokens_seen": 239075328 }, { "epoch": 0.07, "learning_rate": 0.0004684553660982949, "loss": 3.3252, "theoretical_loss": 4.2603043031271035, "tokens_seen": 239140864 }, { "epoch": 0.07, "objective/train/docs_used": 147039, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.313131332397461, "objective/train/theoretical_loss": 4.260162263459744, "objective/train/tokens_used": 259666400, "theoretical_loss": 4.260162263459744, "tokens_seen": 239206400 }, { "epoch": 0.07, "learning_rate": 0.0004684453360080241, "loss": 3.0978, "theoretical_loss": 4.260162263459744, "tokens_seen": 239206400 }, { "epoch": 0.07, "learning_rate": 0.00046843530591775327, "loss": 2.9198, "theoretical_loss": 4.260020273594824, "tokens_seen": 239271936 }, { "epoch": 0.07, "learning_rate": 0.00046842527582748245, "loss": 3.025, "theoretical_loss": 4.259878333501247, "tokens_seen": 239337472 }, { "epoch": 0.07, "learning_rate": 0.0004684152457372117, "loss": 3.0117, "theoretical_loss": 4.259736443147946, "tokens_seen": 239403008 }, { "epoch": 0.07, "learning_rate": 0.0004684052156469408, "loss": 3.3059, "theoretical_loss": 4.259594602503881, "tokens_seen": 239468544 }, { "epoch": 0.07, "learning_rate": 0.00046839518555667005, "loss": 3.4276, "theoretical_loss": 4.259452811538041, "tokens_seen": 239534080 }, { "epoch": 0.07, "learning_rate": 0.00046838515546639923, "loss": 3.3361, "theoretical_loss": 4.259311070219441, "tokens_seen": 239599616 }, { "epoch": 0.07, "learning_rate": 0.0004683751253761284, "loss": 3.3203, "theoretical_loss": 4.259169378517125, "tokens_seen": 239665152 }, { "epoch": 0.07, "learning_rate": 0.0004683650952858576, "loss": 3.2315, "theoretical_loss": 4.259027736400165, "tokens_seen": 239730688 }, { "epoch": 0.07, "learning_rate": 0.0004683550651955868, "loss": 3.014, "theoretical_loss": 4.258886143837661, "tokens_seen": 239796224 }, { "epoch": 0.07, "learning_rate": 0.00046834503510531595, "loss": 3.2752, "theoretical_loss": 4.258744600798739, "tokens_seen": 239861760 }, { "epoch": 0.07, "learning_rate": 0.0004683350050150452, "loss": 3.4321, "theoretical_loss": 4.2586031072525525, "tokens_seen": 239927296 }, { "epoch": 0.07, "learning_rate": 0.0004683249749247743, "loss": 2.9211, "theoretical_loss": 4.258461663168285, "tokens_seen": 239992832 }, { "epoch": 0.07, "learning_rate": 0.00046831494483450355, "loss": 3.0035, "theoretical_loss": 4.258320268515147, "tokens_seen": 240058368 }, { "epoch": 0.07, "learning_rate": 0.0004683049147442327, "loss": 2.9131, "theoretical_loss": 4.258178923262376, "tokens_seen": 240123904 }, { "epoch": 0.07, "learning_rate": 0.0004682948846539619, "loss": 3.3793, "theoretical_loss": 4.258037627379235, "tokens_seen": 240189440 }, { "epoch": 0.07, "learning_rate": 0.0004682848545636911, "loss": 3.182, "theoretical_loss": 4.257896380835018, "tokens_seen": 240254976 }, { "epoch": 0.07, "learning_rate": 0.0004682748244734203, "loss": 3.0911, "theoretical_loss": 4.257755183599045, "tokens_seen": 240320512 }, { "epoch": 0.07, "learning_rate": 0.00046826479438314946, "loss": 3.1443, "theoretical_loss": 4.257614035640662, "tokens_seen": 240386048 }, { "epoch": 0.07, "learning_rate": 0.00046825476429287864, "loss": 3.3931, "theoretical_loss": 4.257472936929246, "tokens_seen": 240451584 }, { "epoch": 0.07, "learning_rate": 0.0004682447342026078, "loss": 3.2777, "theoretical_loss": 4.257331887434198, "tokens_seen": 240517120 }, { "epoch": 0.07, "learning_rate": 0.00046823470411233706, "loss": 3.3189, "theoretical_loss": 4.257190887124946, "tokens_seen": 240582656 }, { "epoch": 0.07, "learning_rate": 0.0004682246740220662, "loss": 3.3554, "theoretical_loss": 4.25704993597095, "tokens_seen": 240648192 }, { "epoch": 0.07, "learning_rate": 0.0004682146439317954, "loss": 3.1868, "theoretical_loss": 4.256909033941691, "tokens_seen": 240713728 }, { "epoch": 0.07, "learning_rate": 0.0004682046138415246, "loss": 3.3277, "theoretical_loss": 4.256768181006683, "tokens_seen": 240779264 }, { "epoch": 0.07, "objective/train/docs_used": 148118, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0634944438934326, "objective/train/theoretical_loss": 4.2566273771354615, "objective/train/tokens_used": 261304800, "theoretical_loss": 4.2566273771354615, "tokens_seen": 240844800 }, { "epoch": 0.07, "learning_rate": 0.0004681945837512538, "loss": 3.1663, "theoretical_loss": 4.2566273771354615, "tokens_seen": 240844800 }, { "epoch": 0.07, "learning_rate": 0.00046818455366098296, "loss": 3.3725, "theoretical_loss": 4.256486622297595, "tokens_seen": 240910336 }, { "epoch": 0.07, "learning_rate": 0.00046817452357071214, "loss": 3.4629, "theoretical_loss": 4.256345916462674, "tokens_seen": 240975872 }, { "epoch": 0.07, "learning_rate": 0.0004681644934804413, "loss": 3.216, "theoretical_loss": 4.256205259600321, "tokens_seen": 241041408 }, { "epoch": 0.07, "learning_rate": 0.00046815446339017056, "loss": 3.2409, "theoretical_loss": 4.256064651680182, "tokens_seen": 241106944 }, { "epoch": 0.07, "learning_rate": 0.0004681444332998997, "loss": 3.3066, "theoretical_loss": 4.255924092671931, "tokens_seen": 241172480 }, { "epoch": 0.07, "learning_rate": 0.0004681344032096289, "loss": 3.2428, "theoretical_loss": 4.255783582545269, "tokens_seen": 241238016 }, { "epoch": 0.07, "learning_rate": 0.00046812437311935805, "loss": 3.2391, "theoretical_loss": 4.255643121269924, "tokens_seen": 241303552 }, { "epoch": 0.07, "learning_rate": 0.0004681143430290873, "loss": 3.1302, "theoretical_loss": 4.255502708815651, "tokens_seen": 241369088 }, { "epoch": 0.07, "learning_rate": 0.00046810431293881646, "loss": 3.1376, "theoretical_loss": 4.255362345152234, "tokens_seen": 241434624 }, { "epoch": 0.07, "learning_rate": 0.00046809428284854565, "loss": 3.2276, "theoretical_loss": 4.255222030249479, "tokens_seen": 241500160 }, { "epoch": 0.07, "learning_rate": 0.0004680842527582748, "loss": 3.1871, "theoretical_loss": 4.255081764077224, "tokens_seen": 241565696 }, { "epoch": 0.07, "learning_rate": 0.000468074222668004, "loss": 3.02, "theoretical_loss": 4.25494154660533, "tokens_seen": 241631232 }, { "epoch": 0.07, "learning_rate": 0.0004680641925777332, "loss": 3.2228, "theoretical_loss": 4.254801377803689, "tokens_seen": 241696768 }, { "epoch": 0.07, "learning_rate": 0.0004680541624874624, "loss": 3.198, "theoretical_loss": 4.254661257642215, "tokens_seen": 241762304 }, { "epoch": 0.07, "learning_rate": 0.00046804413239719155, "loss": 3.2133, "theoretical_loss": 4.254521186090852, "tokens_seen": 241827840 }, { "epoch": 0.07, "learning_rate": 0.0004680341023069208, "loss": 3.2564, "theoretical_loss": 4.254381163119568, "tokens_seen": 241893376 }, { "epoch": 0.07, "learning_rate": 0.00046802407221664997, "loss": 3.1465, "theoretical_loss": 4.254241188698361, "tokens_seen": 241958912 }, { "epoch": 0.07, "learning_rate": 0.00046801404212637915, "loss": 3.3137, "theoretical_loss": 4.2541012627972545, "tokens_seen": 242024448 }, { "epoch": 0.07, "learning_rate": 0.00046800401203610833, "loss": 3.148, "theoretical_loss": 4.2539613853862965, "tokens_seen": 242089984 }, { "epoch": 0.07, "learning_rate": 0.0004679939819458375, "loss": 3.1473, "theoretical_loss": 4.253821556435565, "tokens_seen": 242155520 }, { "epoch": 0.07, "learning_rate": 0.0004679839518555667, "loss": 2.9928, "theoretical_loss": 4.253681775915161, "tokens_seen": 242221056 }, { "epoch": 0.07, "learning_rate": 0.00046797392176529593, "loss": 3.2938, "theoretical_loss": 4.253542043795215, "tokens_seen": 242286592 }, { "epoch": 0.07, "learning_rate": 0.00046796389167502505, "loss": 3.1909, "theoretical_loss": 4.253402360045882, "tokens_seen": 242352128 }, { "epoch": 0.07, "learning_rate": 0.0004679538615847543, "loss": 3.3872, "theoretical_loss": 4.253262724637346, "tokens_seen": 242417664 }, { "epoch": 0.07, "objective/train/docs_used": 148826, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4665045738220215, "objective/train/theoretical_loss": 4.253123137539814, "objective/train/tokens_used": 262943200, "theoretical_loss": 4.253123137539814, "tokens_seen": 242483200 }, { "epoch": 0.07, "learning_rate": 0.0004679438314944834, "loss": 3.1621, "theoretical_loss": 4.253123137539814, "tokens_seen": 242483200 }, { "epoch": 0.07, "learning_rate": 0.00046793380140421265, "loss": 2.9992, "theoretical_loss": 4.252983598723521, "tokens_seen": 242548736 }, { "epoch": 0.07, "learning_rate": 0.00046792377131394183, "loss": 3.2264, "theoretical_loss": 4.25284410815873, "tokens_seen": 242614272 }, { "epoch": 0.07, "learning_rate": 0.000467913741223671, "loss": 3.0202, "theoretical_loss": 4.2527046658157275, "tokens_seen": 242679808 }, { "epoch": 0.07, "learning_rate": 0.0004679037111334002, "loss": 3.2755, "theoretical_loss": 4.252565271664828, "tokens_seen": 242745344 }, { "epoch": 0.07, "learning_rate": 0.00046789368104312943, "loss": 3.2883, "theoretical_loss": 4.252425925676373, "tokens_seen": 242810880 }, { "epoch": 0.07, "learning_rate": 0.00046788365095285856, "loss": 2.821, "theoretical_loss": 4.252286627820727, "tokens_seen": 242876416 }, { "epoch": 0.07, "learning_rate": 0.0004678736208625878, "loss": 3.4768, "theoretical_loss": 4.252147378068285, "tokens_seen": 242941952 }, { "epoch": 0.07, "learning_rate": 0.0004678635907723169, "loss": 3.2938, "theoretical_loss": 4.252008176389465, "tokens_seen": 243007488 }, { "epoch": 0.07, "learning_rate": 0.00046785356068204616, "loss": 3.0992, "theoretical_loss": 4.251869022754712, "tokens_seen": 243073024 }, { "epoch": 0.07, "learning_rate": 0.00046784353059177534, "loss": 3.1465, "theoretical_loss": 4.251729917134498, "tokens_seen": 243138560 }, { "epoch": 0.07, "learning_rate": 0.0004678335005015045, "loss": 3.3141, "theoretical_loss": 4.251590859499322, "tokens_seen": 243204096 }, { "epoch": 0.07, "learning_rate": 0.0004678234704112337, "loss": 3.0572, "theoretical_loss": 4.251451849819704, "tokens_seen": 243269632 }, { "epoch": 0.07, "learning_rate": 0.0004678134403209629, "loss": 3.1641, "theoretical_loss": 4.251312888066197, "tokens_seen": 243335168 }, { "epoch": 0.07, "learning_rate": 0.00046780341023069206, "loss": 2.8932, "theoretical_loss": 4.251173974209375, "tokens_seen": 243400704 }, { "epoch": 0.07, "learning_rate": 0.0004677933801404213, "loss": 2.9821, "theoretical_loss": 4.251035108219839, "tokens_seen": 243466240 }, { "epoch": 0.07, "learning_rate": 0.0004677833500501504, "loss": 3.248, "theoretical_loss": 4.250896290068218, "tokens_seen": 243531776 }, { "epoch": 0.07, "learning_rate": 0.00046777331995987966, "loss": 3.1376, "theoretical_loss": 4.250757519725165, "tokens_seen": 243597312 }, { "epoch": 0.07, "learning_rate": 0.0004677632898696088, "loss": 3.236, "theoretical_loss": 4.25061879716136, "tokens_seen": 243662848 }, { "epoch": 0.07, "learning_rate": 0.000467753259779338, "loss": 2.9136, "theoretical_loss": 4.250480122347507, "tokens_seen": 243728384 }, { "epoch": 0.07, "learning_rate": 0.0004677432296890672, "loss": 2.9435, "theoretical_loss": 4.250341495254337, "tokens_seen": 243793920 }, { "epoch": 0.07, "learning_rate": 0.0004677331995987964, "loss": 2.9687, "theoretical_loss": 4.250202915852608, "tokens_seen": 243859456 }, { "epoch": 0.07, "learning_rate": 0.00046772316950852556, "loss": 3.1614, "theoretical_loss": 4.250064384113102, "tokens_seen": 243924992 }, { "epoch": 0.07, "learning_rate": 0.0004677131394182548, "loss": 2.939, "theoretical_loss": 4.249925900006627, "tokens_seen": 243990528 }, { "epoch": 0.07, "learning_rate": 0.000467703109327984, "loss": 3.036, "theoretical_loss": 4.249787463504019, "tokens_seen": 244056064 }, { "epoch": 0.07, "objective/train/docs_used": 150000, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0193285942077637, "objective/train/theoretical_loss": 4.249649074576134, "objective/train/tokens_used": 264581600, "theoretical_loss": 4.249649074576134, "tokens_seen": 244121600 }, { "epoch": 0.07, "learning_rate": 0.00046769307923771316, "loss": 3.1174, "theoretical_loss": 4.249649074576134, "tokens_seen": 244121600 }, { "epoch": 0.07, "learning_rate": 0.00046768304914744234, "loss": 3.278, "theoretical_loss": 4.249510733193862, "tokens_seen": 244187136 }, { "epoch": 0.07, "learning_rate": 0.0004676730190571715, "loss": 2.9943, "theoretical_loss": 4.249372439328111, "tokens_seen": 244252672 }, { "epoch": 0.07, "learning_rate": 0.00046766298896690076, "loss": 3.3433, "theoretical_loss": 4.249234192949818, "tokens_seen": 244318208 }, { "epoch": 0.07, "learning_rate": 0.0004676529588766299, "loss": 3.2905, "theoretical_loss": 4.249095994029947, "tokens_seen": 244383744 }, { "epoch": 0.07, "learning_rate": 0.0004676429287863591, "loss": 3.0341, "theoretical_loss": 4.248957842539484, "tokens_seen": 244449280 }, { "epoch": 0.07, "learning_rate": 0.00046763289869608825, "loss": 3.3165, "theoretical_loss": 4.248819738449442, "tokens_seen": 244514816 }, { "epoch": 0.07, "learning_rate": 0.0004676228686058175, "loss": 3.0861, "theoretical_loss": 4.2486816817308615, "tokens_seen": 244580352 }, { "epoch": 0.07, "learning_rate": 0.00046761283851554666, "loss": 3.2345, "theoretical_loss": 4.248543672354805, "tokens_seen": 244645888 }, { "epoch": 0.07, "learning_rate": 0.00046760280842527585, "loss": 3.2232, "theoretical_loss": 4.248405710292364, "tokens_seen": 244711424 }, { "epoch": 0.07, "learning_rate": 0.000467592778335005, "loss": 3.1468, "theoretical_loss": 4.248267795514652, "tokens_seen": 244776960 }, { "epoch": 0.07, "learning_rate": 0.0004675827482447342, "loss": 3.122, "theoretical_loss": 4.248129927992808, "tokens_seen": 244842496 }, { "epoch": 0.07, "learning_rate": 0.0004675727181544634, "loss": 3.1999, "theoretical_loss": 4.247992107698002, "tokens_seen": 244908032 }, { "epoch": 0.07, "learning_rate": 0.0004675626880641926, "loss": 3.1818, "theoretical_loss": 4.247854334601421, "tokens_seen": 244973568 }, { "epoch": 0.07, "learning_rate": 0.00046755265797392175, "loss": 3.2209, "theoretical_loss": 4.247716608674283, "tokens_seen": 245039104 }, { "epoch": 0.07, "learning_rate": 0.000467542627883651, "loss": 3.1481, "theoretical_loss": 4.247578929887829, "tokens_seen": 245104640 }, { "epoch": 0.07, "learning_rate": 0.00046753259779338017, "loss": 3.0031, "theoretical_loss": 4.247441298213326, "tokens_seen": 245170176 }, { "epoch": 0.07, "learning_rate": 0.00046752256770310935, "loss": 2.998, "theoretical_loss": 4.247303713622067, "tokens_seen": 245235712 }, { "epoch": 0.07, "learning_rate": 0.00046751253761283853, "loss": 3.1993, "theoretical_loss": 4.247166176085367, "tokens_seen": 245301248 }, { "epoch": 0.07, "learning_rate": 0.0004675025075225677, "loss": 3.1806, "theoretical_loss": 4.247028685574569, "tokens_seen": 245366784 }, { "epoch": 0.07, "learning_rate": 0.0004674924774322969, "loss": 3.3216, "theoretical_loss": 4.246891242061041, "tokens_seen": 245432320 }, { "epoch": 0.07, "learning_rate": 0.00046748244734202613, "loss": 3.0241, "theoretical_loss": 4.246753845516174, "tokens_seen": 245497856 }, { "epoch": 0.07, "learning_rate": 0.00046747241725175525, "loss": 3.3789, "theoretical_loss": 4.246616495911388, "tokens_seen": 245563392 }, { "epoch": 0.07, "learning_rate": 0.0004674623871614845, "loss": 3.1574, "theoretical_loss": 4.246479193218123, "tokens_seen": 245628928 }, { "epoch": 0.07, "learning_rate": 0.0004674523570712136, "loss": 3.1504, "theoretical_loss": 4.246341937407848, "tokens_seen": 245694464 }, { "epoch": 0.07, "objective/train/docs_used": 150657, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3028979301452637, "objective/train/theoretical_loss": 4.246204728452055, "objective/train/tokens_used": 266220000, "theoretical_loss": 4.246204728452055, "tokens_seen": 245760000 }, { "epoch": 0.07, "learning_rate": 0.00046744232698094285, "loss": 3.3196, "theoretical_loss": 4.246204728452055, "tokens_seen": 245760000 }, { "epoch": 0.07, "learning_rate": 0.00046743229689067203, "loss": 3.2711, "theoretical_loss": 4.246067566322259, "tokens_seen": 245825536 }, { "epoch": 0.07, "learning_rate": 0.0004674222668004012, "loss": 3.3492, "theoretical_loss": 4.245930450990007, "tokens_seen": 245891072 }, { "epoch": 0.07, "learning_rate": 0.0004674122367101304, "loss": 3.2023, "theoretical_loss": 4.245793382426861, "tokens_seen": 245956608 }, { "epoch": 0.07, "learning_rate": 0.00046740220661985963, "loss": 3.0581, "theoretical_loss": 4.245656360604417, "tokens_seen": 246022144 }, { "epoch": 0.07, "learning_rate": 0.00046739217652958876, "loss": 3.2156, "theoretical_loss": 4.24551938549429, "tokens_seen": 246087680 }, { "epoch": 0.07, "learning_rate": 0.000467382146439318, "loss": 3.1543, "theoretical_loss": 4.2453824570681205, "tokens_seen": 246153216 }, { "epoch": 0.07, "learning_rate": 0.0004673721163490471, "loss": 2.9836, "theoretical_loss": 4.245245575297577, "tokens_seen": 246218752 }, { "epoch": 0.07, "learning_rate": 0.00046736208625877636, "loss": 3.0194, "theoretical_loss": 4.2451087401543495, "tokens_seen": 246284288 }, { "epoch": 0.07, "learning_rate": 0.00046735205616850554, "loss": 3.1133, "theoretical_loss": 4.244971951610154, "tokens_seen": 246349824 }, { "epoch": 0.07, "learning_rate": 0.0004673420260782347, "loss": 3.1389, "theoretical_loss": 4.24483520963673, "tokens_seen": 246415360 }, { "epoch": 0.07, "learning_rate": 0.0004673319959879639, "loss": 3.3422, "theoretical_loss": 4.244698514205844, "tokens_seen": 246480896 }, { "epoch": 0.07, "learning_rate": 0.0004673219658976931, "loss": 3.0803, "theoretical_loss": 4.244561865289285, "tokens_seen": 246546432 }, { "epoch": 0.07, "learning_rate": 0.00046731193580742226, "loss": 3.1255, "theoretical_loss": 4.244425262858867, "tokens_seen": 246611968 }, { "epoch": 0.07, "learning_rate": 0.0004673019057171515, "loss": 3.152, "theoretical_loss": 4.2442887068864295, "tokens_seen": 246677504 }, { "epoch": 0.07, "learning_rate": 0.0004672918756268806, "loss": 3.2079, "theoretical_loss": 4.244152197343835, "tokens_seen": 246743040 }, { "epoch": 0.07, "learning_rate": 0.00046728184553660986, "loss": 3.4062, "theoretical_loss": 4.244015734202973, "tokens_seen": 246808576 }, { "epoch": 0.07, "learning_rate": 0.000467271815446339, "loss": 3.1603, "theoretical_loss": 4.243879317435755, "tokens_seen": 246874112 }, { "epoch": 0.07, "learning_rate": 0.0004672617853560682, "loss": 3.1124, "theoretical_loss": 4.243742947014117, "tokens_seen": 246939648 }, { "epoch": 0.07, "learning_rate": 0.0004672517552657974, "loss": 2.9212, "theoretical_loss": 4.243606622910021, "tokens_seen": 247005184 }, { "epoch": 0.07, "learning_rate": 0.0004672417251755266, "loss": 3.074, "theoretical_loss": 4.243470345095453, "tokens_seen": 247070720 }, { "epoch": 0.07, "learning_rate": 0.00046723169508525576, "loss": 3.1613, "theoretical_loss": 4.2433341135424225, "tokens_seen": 247136256 }, { "epoch": 0.07, "learning_rate": 0.000467221664994985, "loss": 3.1584, "theoretical_loss": 4.243197928222964, "tokens_seen": 247201792 }, { "epoch": 0.07, "learning_rate": 0.0004672116349047141, "loss": 3.2257, "theoretical_loss": 4.243061789109136, "tokens_seen": 247267328 }, { "epoch": 0.07, "learning_rate": 0.00046720160481444336, "loss": 3.1837, "theoretical_loss": 4.242925696173021, "tokens_seen": 247332864 }, { "epoch": 0.07, "objective/train/docs_used": 151862, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.255047082901001, "objective/train/theoretical_loss": 4.2427896493867285, "objective/train/tokens_used": 267858400, "theoretical_loss": 4.2427896493867285, "tokens_seen": 247398400 }, { "epoch": 0.07, "learning_rate": 0.0004671915747241725, "loss": 2.9149, "theoretical_loss": 4.2427896493867285, "tokens_seen": 247398400 }, { "epoch": 0.07, "learning_rate": 0.0004671815446339017, "loss": 2.8987, "theoretical_loss": 4.242653648722387, "tokens_seen": 247463936 }, { "epoch": 0.08, "learning_rate": 0.0004671715145436309, "loss": 2.9504, "theoretical_loss": 4.242517694152154, "tokens_seen": 247529472 }, { "epoch": 0.08, "learning_rate": 0.0004671614844533601, "loss": 3.1881, "theoretical_loss": 4.24238178564821, "tokens_seen": 247595008 }, { "epoch": 0.08, "learning_rate": 0.00046715145436308927, "loss": 2.9594, "theoretical_loss": 4.242245923182756, "tokens_seen": 247660544 }, { "epoch": 0.08, "learning_rate": 0.00046714142427281845, "loss": 3.1327, "theoretical_loss": 4.242110106728022, "tokens_seen": 247726080 }, { "epoch": 0.08, "learning_rate": 0.00046713139418254763, "loss": 3.0429, "theoretical_loss": 4.241974336256261, "tokens_seen": 247791616 }, { "epoch": 0.08, "learning_rate": 0.00046712136409227686, "loss": 3.1163, "theoretical_loss": 4.241838611739748, "tokens_seen": 247857152 }, { "epoch": 0.08, "learning_rate": 0.000467111334002006, "loss": 3.2067, "theoretical_loss": 4.241702933150783, "tokens_seen": 247922688 }, { "epoch": 0.08, "learning_rate": 0.0004671013039117352, "loss": 3.3482, "theoretical_loss": 4.241567300461693, "tokens_seen": 247988224 }, { "epoch": 0.08, "learning_rate": 0.00046709127382146435, "loss": 3.3143, "theoretical_loss": 4.241431713644823, "tokens_seen": 248053760 }, { "epoch": 0.08, "learning_rate": 0.0004670812437311936, "loss": 3.2176, "theoretical_loss": 4.241296172672547, "tokens_seen": 248119296 }, { "epoch": 0.08, "learning_rate": 0.00046707121364092277, "loss": 3.1752, "theoretical_loss": 4.24116067751726, "tokens_seen": 248184832 }, { "epoch": 0.08, "learning_rate": 0.00046706118355065195, "loss": 3.2537, "theoretical_loss": 4.241025228151383, "tokens_seen": 248250368 }, { "epoch": 0.08, "learning_rate": 0.00046705115346038113, "loss": 3.4286, "theoretical_loss": 4.24088982454736, "tokens_seen": 248315904 }, { "epoch": 0.08, "learning_rate": 0.00046704112337011037, "loss": 3.248, "theoretical_loss": 4.240754466677659, "tokens_seen": 248381440 }, { "epoch": 0.08, "learning_rate": 0.0004670310932798395, "loss": 2.9887, "theoretical_loss": 4.240619154514771, "tokens_seen": 248446976 }, { "epoch": 0.08, "learning_rate": 0.00046702106318956873, "loss": 3.2964, "theoretical_loss": 4.240483888031212, "tokens_seen": 248512512 }, { "epoch": 0.08, "learning_rate": 0.00046701103309929786, "loss": 2.9216, "theoretical_loss": 4.240348667199521, "tokens_seen": 248578048 }, { "epoch": 0.08, "learning_rate": 0.0004670010030090271, "loss": 2.9462, "theoretical_loss": 4.240213491992261, "tokens_seen": 248643584 }, { "epoch": 0.08, "learning_rate": 0.0004669909729187563, "loss": 3.2691, "theoretical_loss": 4.240078362382019, "tokens_seen": 248709120 }, { "epoch": 0.08, "learning_rate": 0.00046698094282848545, "loss": 2.8617, "theoretical_loss": 4.239943278341404, "tokens_seen": 248774656 }, { "epoch": 0.08, "learning_rate": 0.00046697091273821464, "loss": 2.9596, "theoretical_loss": 4.239808239843052, "tokens_seen": 248840192 }, { "epoch": 0.08, "learning_rate": 0.0004669608826479438, "loss": 2.9612, "theoretical_loss": 4.239673246859619, "tokens_seen": 248905728 }, { "epoch": 0.08, "learning_rate": 0.00046695085255767305, "loss": 2.9767, "theoretical_loss": 4.239538299363788, "tokens_seen": 248971264 }, { "epoch": 0.08, "objective/train/docs_used": 152583, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3142855167388916, "objective/train/theoretical_loss": 4.239403397328261, "objective/train/tokens_used": 269496800, "theoretical_loss": 4.239403397328261, "tokens_seen": 249036800 }, { "epoch": 0.08, "learning_rate": 0.00046694082246740223, "loss": 3.3554, "theoretical_loss": 4.239403397328261, "tokens_seen": 249036800 }, { "epoch": 0.08, "learning_rate": 0.0004669307923771314, "loss": 3.1367, "theoretical_loss": 4.239268540725769, "tokens_seen": 249102336 }, { "epoch": 0.08, "learning_rate": 0.0004669207622868606, "loss": 3.0328, "theoretical_loss": 4.239133729529064, "tokens_seen": 249167872 }, { "epoch": 0.08, "learning_rate": 0.00046691073219658983, "loss": 3.1881, "theoretical_loss": 4.2389989637109196, "tokens_seen": 249233408 }, { "epoch": 0.08, "learning_rate": 0.00046690070210631896, "loss": 3.1403, "theoretical_loss": 4.2388642432441355, "tokens_seen": 249298944 }, { "epoch": 0.08, "learning_rate": 0.0004668906720160482, "loss": 3.099, "theoretical_loss": 4.238729568101535, "tokens_seen": 249364480 }, { "epoch": 0.08, "learning_rate": 0.0004668806419257773, "loss": 3.2886, "theoretical_loss": 4.238594938255963, "tokens_seen": 249430016 }, { "epoch": 0.08, "learning_rate": 0.00046687061183550656, "loss": 2.9976, "theoretical_loss": 4.2384603536802885, "tokens_seen": 249495552 }, { "epoch": 0.08, "learning_rate": 0.00046686058174523574, "loss": 3.1615, "theoretical_loss": 4.238325814347404, "tokens_seen": 249561088 }, { "epoch": 0.08, "learning_rate": 0.0004668505516549649, "loss": 3.1596, "theoretical_loss": 4.238191320230227, "tokens_seen": 249626624 }, { "epoch": 0.08, "learning_rate": 0.0004668405215646941, "loss": 3.2094, "theoretical_loss": 4.238056871301695, "tokens_seen": 249692160 }, { "epoch": 0.08, "learning_rate": 0.0004668304914744233, "loss": 3.0536, "theoretical_loss": 4.237922467534771, "tokens_seen": 249757696 }, { "epoch": 0.08, "learning_rate": 0.00046682046138415246, "loss": 2.9296, "theoretical_loss": 4.237788108902441, "tokens_seen": 249823232 }, { "epoch": 0.08, "learning_rate": 0.0004668104312938817, "loss": 3.278, "theoretical_loss": 4.237653795377714, "tokens_seen": 249888768 }, { "epoch": 0.08, "learning_rate": 0.0004668004012036108, "loss": 3.0784, "theoretical_loss": 4.237519526933622, "tokens_seen": 249954304 }, { "epoch": 0.08, "learning_rate": 0.00046679037111334006, "loss": 3.3284, "theoretical_loss": 4.2373853035432205, "tokens_seen": 250019840 }, { "epoch": 0.08, "learning_rate": 0.0004667803410230692, "loss": 3.1476, "theoretical_loss": 4.237251125179588, "tokens_seen": 250085376 }, { "epoch": 0.08, "learning_rate": 0.0004667703109327984, "loss": 3.4263, "theoretical_loss": 4.237116991815826, "tokens_seen": 250150912 }, { "epoch": 0.08, "learning_rate": 0.0004667602808425276, "loss": 3.3711, "theoretical_loss": 4.23698290342506, "tokens_seen": 250216448 }, { "epoch": 0.08, "learning_rate": 0.0004667502507522568, "loss": 3.1352, "theoretical_loss": 4.236848859980437, "tokens_seen": 250281984 }, { "epoch": 0.08, "learning_rate": 0.00046674022066198596, "loss": 3.1904, "theoretical_loss": 4.23671486145513, "tokens_seen": 250347520 }, { "epoch": 0.08, "learning_rate": 0.0004667301905717152, "loss": 3.1398, "theoretical_loss": 4.236580907822331, "tokens_seen": 250413056 }, { "epoch": 0.08, "learning_rate": 0.0004667201604814443, "loss": 3.2989, "theoretical_loss": 4.236446999055257, "tokens_seen": 250478592 }, { "epoch": 0.08, "learning_rate": 0.00046671013039117356, "loss": 3.215, "theoretical_loss": 4.2363131351271495, "tokens_seen": 250544128 }, { "epoch": 0.08, "learning_rate": 0.0004667001003009027, "loss": 3.0471, "theoretical_loss": 4.2361793160112695, "tokens_seen": 250609664 }, { "epoch": 0.08, "objective/train/docs_used": 153242, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2777795791625977, "objective/train/theoretical_loss": 4.236045541680905, "objective/train/tokens_used": 271135200, "theoretical_loss": 4.236045541680905, "tokens_seen": 250675200 }, { "epoch": 0.08, "learning_rate": 0.0004666900702106319, "loss": 3.1679, "theoretical_loss": 4.236045541680905, "tokens_seen": 250675200 }, { "epoch": 0.08, "learning_rate": 0.0004666800401203611, "loss": 3.0121, "theoretical_loss": 4.235911812109363, "tokens_seen": 250740736 }, { "epoch": 0.08, "learning_rate": 0.0004666700100300903, "loss": 3.1682, "theoretical_loss": 4.235778127269976, "tokens_seen": 250806272 }, { "epoch": 0.08, "learning_rate": 0.00046665997993981947, "loss": 3.236, "theoretical_loss": 4.235644487136098, "tokens_seen": 250871808 }, { "epoch": 0.08, "learning_rate": 0.00046664994984954865, "loss": 3.1782, "theoretical_loss": 4.235510891681108, "tokens_seen": 250937344 }, { "epoch": 0.08, "learning_rate": 0.00046663991975927783, "loss": 2.8634, "theoretical_loss": 4.235377340878404, "tokens_seen": 251002880 }, { "epoch": 0.08, "learning_rate": 0.00046662988966900706, "loss": 3.3324, "theoretical_loss": 4.23524383470141, "tokens_seen": 251068416 }, { "epoch": 0.08, "learning_rate": 0.0004666198595787362, "loss": 3.346, "theoretical_loss": 4.235110373123572, "tokens_seen": 251133952 }, { "epoch": 0.08, "learning_rate": 0.00046660982948846543, "loss": 3.2462, "theoretical_loss": 4.2349769561183574, "tokens_seen": 251199488 }, { "epoch": 0.08, "learning_rate": 0.00046659979939819455, "loss": 3.1068, "theoretical_loss": 4.2348435836592575, "tokens_seen": 251265024 }, { "epoch": 0.08, "learning_rate": 0.0004665897693079238, "loss": 3.1695, "theoretical_loss": 4.234710255719786, "tokens_seen": 251330560 }, { "epoch": 0.08, "learning_rate": 0.00046657973921765297, "loss": 3.1471, "theoretical_loss": 4.234576972273481, "tokens_seen": 251396096 }, { "epoch": 0.08, "learning_rate": 0.00046656970912738215, "loss": 3.0517, "theoretical_loss": 4.234443733293899, "tokens_seen": 251461632 }, { "epoch": 0.08, "learning_rate": 0.00046655967903711133, "loss": 3.1296, "theoretical_loss": 4.234310538754624, "tokens_seen": 251527168 }, { "epoch": 0.08, "learning_rate": 0.00046654964894684057, "loss": 3.2316, "theoretical_loss": 4.2341773886292575, "tokens_seen": 251592704 }, { "epoch": 0.08, "learning_rate": 0.0004665396188565697, "loss": 3.0754, "theoretical_loss": 4.234044282891429, "tokens_seen": 251658240 }, { "epoch": 0.08, "learning_rate": 0.00046652958876629893, "loss": 3.4376, "theoretical_loss": 4.233911221514787, "tokens_seen": 251723776 }, { "epoch": 0.08, "learning_rate": 0.00046651955867602806, "loss": 3.1209, "theoretical_loss": 4.233778204473002, "tokens_seen": 251789312 }, { "epoch": 0.08, "learning_rate": 0.0004665095285857573, "loss": 3.054, "theoretical_loss": 4.23364523173977, "tokens_seen": 251854848 }, { "epoch": 0.08, "learning_rate": 0.0004664994984954865, "loss": 3.1497, "theoretical_loss": 4.233512303288807, "tokens_seen": 251920384 }, { "epoch": 0.08, "learning_rate": 0.00046648946840521565, "loss": 3.243, "theoretical_loss": 4.233379419093851, "tokens_seen": 251985920 }, { "epoch": 0.08, "learning_rate": 0.00046647943831494484, "loss": 2.9777, "theoretical_loss": 4.233246579128666, "tokens_seen": 252051456 }, { "epoch": 0.08, "learning_rate": 0.000466469408224674, "loss": 3.0244, "theoretical_loss": 4.233113783367033, "tokens_seen": 252116992 }, { "epoch": 0.08, "learning_rate": 0.0004664593781344032, "loss": 2.8659, "theoretical_loss": 4.232981031782761, "tokens_seen": 252182528 }, { "epoch": 0.08, "learning_rate": 0.00046644934804413243, "loss": 3.0599, "theoretical_loss": 4.232848324349677, "tokens_seen": 252248064 }, { "epoch": 0.08, "objective/train/docs_used": 154565, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2938530445098877, "objective/train/theoretical_loss": 4.232715661041632, "objective/train/tokens_used": 272773600, "theoretical_loss": 4.232715661041632, "tokens_seen": 252313600 }, { "epoch": 0.08, "learning_rate": 0.00046643931795386156, "loss": 3.1579, "theoretical_loss": 4.232715661041632, "tokens_seen": 252313600 }, { "epoch": 0.08, "learning_rate": 0.0004664292878635908, "loss": 2.8977, "theoretical_loss": 4.232583041832499, "tokens_seen": 252379136 }, { "epoch": 0.08, "learning_rate": 0.0004664192577733199, "loss": 2.9725, "theoretical_loss": 4.232450466696174, "tokens_seen": 252444672 }, { "epoch": 0.08, "learning_rate": 0.00046640922768304916, "loss": 3.247, "theoretical_loss": 4.2323179356065745, "tokens_seen": 252510208 }, { "epoch": 0.08, "learning_rate": 0.00046639919759277834, "loss": 3.0137, "theoretical_loss": 4.23218544853764, "tokens_seen": 252575744 }, { "epoch": 0.08, "learning_rate": 0.0004663891675025075, "loss": 3.1024, "theoretical_loss": 4.232053005463333, "tokens_seen": 252641280 }, { "epoch": 0.08, "learning_rate": 0.0004663791374122367, "loss": 3.1656, "theoretical_loss": 4.231920606357638, "tokens_seen": 252706816 }, { "epoch": 0.08, "learning_rate": 0.00046636910732196594, "loss": 3.2556, "theoretical_loss": 4.231788251194559, "tokens_seen": 252772352 }, { "epoch": 0.08, "learning_rate": 0.00046635907723169506, "loss": 3.2828, "theoretical_loss": 4.231655939948127, "tokens_seen": 252837888 }, { "epoch": 0.08, "learning_rate": 0.0004663490471414243, "loss": 3.0292, "theoretical_loss": 4.231523672592392, "tokens_seen": 252903424 }, { "epoch": 0.08, "learning_rate": 0.0004663390170511534, "loss": 3.2209, "theoretical_loss": 4.231391449101425, "tokens_seen": 252968960 }, { "epoch": 0.08, "learning_rate": 0.00046632898696088266, "loss": 3.3026, "theoretical_loss": 4.231259269449322, "tokens_seen": 253034496 }, { "epoch": 0.08, "learning_rate": 0.00046631895687061184, "loss": 3.2373, "theoretical_loss": 4.231127133610198, "tokens_seen": 253100032 }, { "epoch": 0.08, "learning_rate": 0.000466308926780341, "loss": 3.1833, "theoretical_loss": 4.230995041558194, "tokens_seen": 253165568 }, { "epoch": 0.08, "learning_rate": 0.0004662988966900702, "loss": 3.0795, "theoretical_loss": 4.230862993267468, "tokens_seen": 253231104 }, { "epoch": 0.08, "learning_rate": 0.0004662888665997994, "loss": 3.1932, "theoretical_loss": 4.230730988712205, "tokens_seen": 253296640 }, { "epoch": 0.08, "learning_rate": 0.00046627883650952857, "loss": 3.2634, "theoretical_loss": 4.230599027866606, "tokens_seen": 253362176 }, { "epoch": 0.08, "learning_rate": 0.0004662688064192578, "loss": 3.3007, "theoretical_loss": 4.2304671107048994, "tokens_seen": 253427712 }, { "epoch": 0.08, "learning_rate": 0.00046625877632898693, "loss": 3.1446, "theoretical_loss": 4.2303352372013325, "tokens_seen": 253493248 }, { "epoch": 0.08, "learning_rate": 0.00046624874623871616, "loss": 3.2354, "theoretical_loss": 4.230203407330176, "tokens_seen": 253558784 }, { "epoch": 0.08, "learning_rate": 0.00046623871614844535, "loss": 3.2521, "theoretical_loss": 4.230071621065721, "tokens_seen": 253624320 }, { "epoch": 0.08, "learning_rate": 0.0004662286860581745, "loss": 3.3151, "theoretical_loss": 4.2299398783822815, "tokens_seen": 253689856 }, { "epoch": 0.08, "learning_rate": 0.0004662186559679037, "loss": 3.0526, "theoretical_loss": 4.229808179254192, "tokens_seen": 253755392 }, { "epoch": 0.08, "learning_rate": 0.0004662086258776329, "loss": 3.1744, "theoretical_loss": 4.22967652365581, "tokens_seen": 253820928 }, { "epoch": 0.08, "learning_rate": 0.0004661985957873621, "loss": 3.2694, "theoretical_loss": 4.229544911561513, "tokens_seen": 253886464 }, { "epoch": 0.08, "objective/train/docs_used": 156106, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.7020857334136963, "objective/train/theoretical_loss": 4.229413342945703, "objective/train/tokens_used": 274412000, "theoretical_loss": 4.229413342945703, "tokens_seen": 253952000 }, { "epoch": 0.08, "learning_rate": 0.0004661885656970913, "loss": 3.1877, "theoretical_loss": 4.229413342945703, "tokens_seen": 253952000 }, { "epoch": 0.08, "learning_rate": 0.0004661785356068205, "loss": 3.2587, "theoretical_loss": 4.229281817782801, "tokens_seen": 254017536 }, { "epoch": 0.08, "learning_rate": 0.00046616850551654967, "loss": 3.2171, "theoretical_loss": 4.229150336047251, "tokens_seen": 254083072 }, { "epoch": 0.08, "learning_rate": 0.00046615847542627885, "loss": 2.7179, "theoretical_loss": 4.229018897713519, "tokens_seen": 254148608 }, { "epoch": 0.08, "learning_rate": 0.00046614844533600803, "loss": 3.0186, "theoretical_loss": 4.22888750275609, "tokens_seen": 254214144 }, { "epoch": 0.08, "learning_rate": 0.00046613841524573727, "loss": 3.1104, "theoretical_loss": 4.228756151149475, "tokens_seen": 254279680 }, { "epoch": 0.08, "learning_rate": 0.0004661283851554664, "loss": 3.2379, "theoretical_loss": 4.228624842868202, "tokens_seen": 254345216 }, { "epoch": 0.08, "learning_rate": 0.00046611835506519563, "loss": 3.1056, "theoretical_loss": 4.228493577886824, "tokens_seen": 254410752 }, { "epoch": 0.08, "learning_rate": 0.00046610832497492475, "loss": 3.2572, "theoretical_loss": 4.228362356179913, "tokens_seen": 254476288 }, { "epoch": 0.08, "learning_rate": 0.000466098294884654, "loss": 3.019, "theoretical_loss": 4.228231177722063, "tokens_seen": 254541824 }, { "epoch": 0.08, "learning_rate": 0.00046608826479438317, "loss": 3.1823, "theoretical_loss": 4.228100042487892, "tokens_seen": 254607360 }, { "epoch": 0.08, "learning_rate": 0.00046607823470411235, "loss": 3.0422, "theoretical_loss": 4.227968950452035, "tokens_seen": 254672896 }, { "epoch": 0.08, "learning_rate": 0.00046606820461384153, "loss": 3.0621, "theoretical_loss": 4.227837901589153, "tokens_seen": 254738432 }, { "epoch": 0.08, "learning_rate": 0.00046605817452357077, "loss": 2.9792, "theoretical_loss": 4.227706895873924, "tokens_seen": 254803968 }, { "epoch": 0.08, "learning_rate": 0.0004660481444332999, "loss": 3.2174, "theoretical_loss": 4.227575933281051, "tokens_seen": 254869504 }, { "epoch": 0.08, "learning_rate": 0.00046603811434302913, "loss": 3.2159, "theoretical_loss": 4.227445013785257, "tokens_seen": 254935040 }, { "epoch": 0.08, "learning_rate": 0.00046602808425275826, "loss": 3.2537, "theoretical_loss": 4.227314137361285, "tokens_seen": 255000576 }, { "epoch": 0.08, "learning_rate": 0.0004660180541624875, "loss": 3.2481, "theoretical_loss": 4.227183303983901, "tokens_seen": 255066112 }, { "epoch": 0.08, "learning_rate": 0.0004660080240722167, "loss": 3.1691, "theoretical_loss": 4.227052513627893, "tokens_seen": 255131648 }, { "epoch": 0.08, "learning_rate": 0.00046599799398194586, "loss": 3.1694, "theoretical_loss": 4.226921766268067, "tokens_seen": 255197184 }, { "epoch": 0.08, "learning_rate": 0.00046598796389167504, "loss": 3.1596, "theoretical_loss": 4.226791061879253, "tokens_seen": 255262720 }, { "epoch": 0.08, "learning_rate": 0.0004659779338014042, "loss": 3.0473, "theoretical_loss": 4.226660400436302, "tokens_seen": 255328256 }, { "epoch": 0.08, "learning_rate": 0.0004659679037111334, "loss": 3.1521, "theoretical_loss": 4.226529781914084, "tokens_seen": 255393792 }, { "epoch": 0.08, "learning_rate": 0.00046595787362086263, "loss": 3.0574, "theoretical_loss": 4.226399206287493, "tokens_seen": 255459328 }, { "epoch": 0.08, "learning_rate": 0.00046594784353059176, "loss": 3.1197, "theoretical_loss": 4.226268673531442, "tokens_seen": 255524864 }, { "epoch": 0.08, "objective/train/docs_used": 156630, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.661715507507324, "objective/train/theoretical_loss": 4.226138183620867, "objective/train/tokens_used": 276050400, "theoretical_loss": 4.226138183620867, "tokens_seen": 255590400 }, { "epoch": 0.08, "learning_rate": 0.000465937813440321, "loss": 3.0664, "theoretical_loss": 4.226138183620867, "tokens_seen": 255590400 }, { "epoch": 0.08, "learning_rate": 0.0004659277833500501, "loss": 3.0895, "theoretical_loss": 4.226007736530723, "tokens_seen": 255655936 }, { "epoch": 0.08, "learning_rate": 0.00046591775325977936, "loss": 2.9574, "theoretical_loss": 4.225877332235987, "tokens_seen": 255721472 }, { "epoch": 0.08, "learning_rate": 0.00046590772316950854, "loss": 3.1183, "theoretical_loss": 4.225746970711657, "tokens_seen": 255787008 }, { "epoch": 0.08, "learning_rate": 0.0004658976930792377, "loss": 2.9636, "theoretical_loss": 4.225616651932753, "tokens_seen": 255852544 }, { "epoch": 0.08, "learning_rate": 0.0004658876629889669, "loss": 3.2771, "theoretical_loss": 4.225486375874315, "tokens_seen": 255918080 }, { "epoch": 0.08, "learning_rate": 0.00046587763289869614, "loss": 2.8867, "theoretical_loss": 4.225356142511402, "tokens_seen": 255983616 }, { "epoch": 0.08, "learning_rate": 0.00046586760280842526, "loss": 3.2734, "theoretical_loss": 4.225225951819099, "tokens_seen": 256049152 }, { "epoch": 0.08, "learning_rate": 0.0004658575727181545, "loss": 3.1851, "theoretical_loss": 4.225095803772507, "tokens_seen": 256114688 }, { "epoch": 0.08, "learning_rate": 0.0004658475426278836, "loss": 3.0544, "theoretical_loss": 4.22496569834675, "tokens_seen": 256180224 }, { "epoch": 0.08, "learning_rate": 0.00046583751253761286, "loss": 3.1337, "theoretical_loss": 4.224835635516973, "tokens_seen": 256245760 }, { "epoch": 0.08, "learning_rate": 0.00046582748244734204, "loss": 2.9687, "theoretical_loss": 4.224705615258341, "tokens_seen": 256311296 }, { "epoch": 0.08, "learning_rate": 0.0004658174523570712, "loss": 3.1143, "theoretical_loss": 4.224575637546041, "tokens_seen": 256376832 }, { "epoch": 0.08, "learning_rate": 0.0004658074222668004, "loss": 3.1526, "theoretical_loss": 4.224445702355279, "tokens_seen": 256442368 }, { "epoch": 0.08, "learning_rate": 0.0004657973921765296, "loss": 3.3006, "theoretical_loss": 4.2243158096612845, "tokens_seen": 256507904 }, { "epoch": 0.08, "learning_rate": 0.00046578736208625877, "loss": 3.0952, "theoretical_loss": 4.224185959439305, "tokens_seen": 256573440 }, { "epoch": 0.08, "learning_rate": 0.000465777331995988, "loss": 2.8397, "theoretical_loss": 4.22405615166461, "tokens_seen": 256638976 }, { "epoch": 0.08, "learning_rate": 0.00046576730190571713, "loss": 3.1773, "theoretical_loss": 4.22392638631249, "tokens_seen": 256704512 }, { "epoch": 0.08, "learning_rate": 0.00046575727181544636, "loss": 3.0894, "theoretical_loss": 4.223796663358255, "tokens_seen": 256770048 }, { "epoch": 0.08, "learning_rate": 0.00046574724172517555, "loss": 3.044, "theoretical_loss": 4.223666982777237, "tokens_seen": 256835584 }, { "epoch": 0.08, "learning_rate": 0.0004657372116349047, "loss": 3.1196, "theoretical_loss": 4.223537344544788, "tokens_seen": 256901120 }, { "epoch": 0.08, "learning_rate": 0.0004657271815446339, "loss": 3.3258, "theoretical_loss": 4.223407748636282, "tokens_seen": 256966656 }, { "epoch": 0.08, "learning_rate": 0.0004657171514543631, "loss": 3.1867, "theoretical_loss": 4.22327819502711, "tokens_seen": 257032192 }, { "epoch": 0.08, "learning_rate": 0.00046570712136409227, "loss": 3.0392, "theoretical_loss": 4.223148683692687, "tokens_seen": 257097728 }, { "epoch": 0.08, "learning_rate": 0.0004656970912738215, "loss": 3.1276, "theoretical_loss": 4.223019214608446, "tokens_seen": 257163264 }, { "epoch": 0.08, "objective/train/docs_used": 157258, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.008852005004883, "objective/train/theoretical_loss": 4.222889787749845, "objective/train/tokens_used": 277688800, "theoretical_loss": 4.222889787749845, "tokens_seen": 257228800 }, { "epoch": 0.08, "learning_rate": 0.00046568706118355063, "loss": 3.0922, "theoretical_loss": 4.222889787749845, "tokens_seen": 257228800 }, { "epoch": 0.08, "learning_rate": 0.00046567703109327987, "loss": 2.8066, "theoretical_loss": 4.222760403092358, "tokens_seen": 257294336 }, { "epoch": 0.08, "learning_rate": 0.000465667001003009, "loss": 3.0272, "theoretical_loss": 4.22263106061148, "tokens_seen": 257359872 }, { "epoch": 0.08, "learning_rate": 0.00046565697091273823, "loss": 3.0383, "theoretical_loss": 4.222501760282729, "tokens_seen": 257425408 }, { "epoch": 0.08, "learning_rate": 0.0004656469408224674, "loss": 2.9278, "theoretical_loss": 4.22237250208164, "tokens_seen": 257490944 }, { "epoch": 0.08, "learning_rate": 0.0004656369107321966, "loss": 3.0506, "theoretical_loss": 4.222243285983772, "tokens_seen": 257556480 }, { "epoch": 0.08, "learning_rate": 0.0004656268806419258, "loss": 2.754, "theoretical_loss": 4.222114111964703, "tokens_seen": 257622016 }, { "epoch": 0.08, "learning_rate": 0.00046561685055165495, "loss": 2.9813, "theoretical_loss": 4.221984980000029, "tokens_seen": 257687552 }, { "epoch": 0.08, "learning_rate": 0.00046560682046138414, "loss": 2.94, "theoretical_loss": 4.2218558900653695, "tokens_seen": 257753088 }, { "epoch": 0.08, "learning_rate": 0.00046559679037111337, "loss": 2.7958, "theoretical_loss": 4.221726842136364, "tokens_seen": 257818624 }, { "epoch": 0.08, "learning_rate": 0.0004655867602808425, "loss": 3.1393, "theoretical_loss": 4.2215978361886695, "tokens_seen": 257884160 }, { "epoch": 0.08, "learning_rate": 0.00046557673019057173, "loss": 3.1812, "theoretical_loss": 4.221468872197967, "tokens_seen": 257949696 }, { "epoch": 0.08, "learning_rate": 0.0004655667001003009, "loss": 3.066, "theoretical_loss": 4.221339950139956, "tokens_seen": 258015232 }, { "epoch": 0.08, "learning_rate": 0.0004655566700100301, "loss": 2.9906, "theoretical_loss": 4.221211069990357, "tokens_seen": 258080768 }, { "epoch": 0.08, "learning_rate": 0.0004655466399197593, "loss": 3.1732, "theoretical_loss": 4.221082231724908, "tokens_seen": 258146304 }, { "epoch": 0.08, "learning_rate": 0.00046553660982948846, "loss": 3.0643, "theoretical_loss": 4.22095343531937, "tokens_seen": 258211840 }, { "epoch": 0.08, "learning_rate": 0.00046552657973921764, "loss": 3.1475, "theoretical_loss": 4.220824680749525, "tokens_seen": 258277376 }, { "epoch": 0.08, "learning_rate": 0.0004655165496489469, "loss": 3.227, "theoretical_loss": 4.220695967991171, "tokens_seen": 258342912 }, { "epoch": 0.08, "learning_rate": 0.000465506519558676, "loss": 3.1797, "theoretical_loss": 4.220567297020131, "tokens_seen": 258408448 }, { "epoch": 0.08, "learning_rate": 0.00046549648946840524, "loss": 3.0443, "theoretical_loss": 4.220438667812244, "tokens_seen": 258473984 }, { "epoch": 0.08, "learning_rate": 0.00046548645937813436, "loss": 3.0716, "theoretical_loss": 4.220310080343373, "tokens_seen": 258539520 }, { "epoch": 0.08, "learning_rate": 0.0004654764292878636, "loss": 3.2806, "theoretical_loss": 4.220181534589398, "tokens_seen": 258605056 }, { "epoch": 0.08, "learning_rate": 0.0004654663991975928, "loss": 3.0484, "theoretical_loss": 4.22005303052622, "tokens_seen": 258670592 }, { "epoch": 0.08, "learning_rate": 0.00046545636910732196, "loss": 3.2093, "theoretical_loss": 4.219924568129759, "tokens_seen": 258736128 }, { "epoch": 0.08, "learning_rate": 0.0004654463390170512, "loss": 2.9654, "theoretical_loss": 4.219796147375957, "tokens_seen": 258801664 }, { "epoch": 0.08, "objective/train/docs_used": 158510, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9906694889068604, "objective/train/theoretical_loss": 4.219667768240775, "objective/train/tokens_used": 279327200, "theoretical_loss": 4.219667768240775, "tokens_seen": 258867200 }, { "epoch": 0.08, "learning_rate": 0.0004654363089267803, "loss": 3.068, "theoretical_loss": 4.219667768240775, "tokens_seen": 258867200 }, { "epoch": 0.08, "learning_rate": 0.00046542627883650956, "loss": 3.2658, "theoretical_loss": 4.219539430700195, "tokens_seen": 258932736 }, { "epoch": 0.08, "learning_rate": 0.00046541624874623874, "loss": 3.0762, "theoretical_loss": 4.2194111347302155, "tokens_seen": 258998272 }, { "epoch": 0.08, "learning_rate": 0.0004654062186559679, "loss": 3.2659, "theoretical_loss": 4.219282880306859, "tokens_seen": 259063808 }, { "epoch": 0.08, "learning_rate": 0.0004653961885656971, "loss": 3.0452, "theoretical_loss": 4.219154667406166, "tokens_seen": 259129344 }, { "epoch": 0.08, "learning_rate": 0.00046538615847542634, "loss": 3.0682, "theoretical_loss": 4.219026496004198, "tokens_seen": 259194880 }, { "epoch": 0.08, "learning_rate": 0.00046537612838515546, "loss": 3.0652, "theoretical_loss": 4.218898366077035, "tokens_seen": 259260416 }, { "epoch": 0.08, "learning_rate": 0.0004653660982948847, "loss": 3.2532, "theoretical_loss": 4.218770277600775, "tokens_seen": 259325952 }, { "epoch": 0.08, "learning_rate": 0.0004653560682046138, "loss": 2.9702, "theoretical_loss": 4.218642230551541, "tokens_seen": 259391488 }, { "epoch": 0.08, "learning_rate": 0.00046534603811434306, "loss": 2.987, "theoretical_loss": 4.218514224905472, "tokens_seen": 259457024 }, { "epoch": 0.08, "learning_rate": 0.00046533600802407224, "loss": 3.1381, "theoretical_loss": 4.218386260638727, "tokens_seen": 259522560 }, { "epoch": 0.08, "learning_rate": 0.0004653259779338014, "loss": 2.8499, "theoretical_loss": 4.2182583377274865, "tokens_seen": 259588096 }, { "epoch": 0.08, "learning_rate": 0.0004653159478435306, "loss": 3.0011, "theoretical_loss": 4.218130456147948, "tokens_seen": 259653632 }, { "epoch": 0.08, "learning_rate": 0.0004653059177532598, "loss": 3.0006, "theoretical_loss": 4.218002615876332, "tokens_seen": 259719168 }, { "epoch": 0.08, "learning_rate": 0.00046529588766298897, "loss": 2.8484, "theoretical_loss": 4.217874816888877, "tokens_seen": 259784704 }, { "epoch": 0.08, "learning_rate": 0.0004652858575727182, "loss": 3.0114, "theoretical_loss": 4.217747059161839, "tokens_seen": 259850240 }, { "epoch": 0.08, "learning_rate": 0.00046527582748244733, "loss": 3.1379, "theoretical_loss": 4.217619342671498, "tokens_seen": 259915776 }, { "epoch": 0.08, "learning_rate": 0.00046526579739217656, "loss": 3.1485, "theoretical_loss": 4.2174916673941505, "tokens_seen": 259981312 }, { "epoch": 0.08, "learning_rate": 0.00046525576730190575, "loss": 3.1037, "theoretical_loss": 4.217364033306113, "tokens_seen": 260046848 }, { "epoch": 0.08, "learning_rate": 0.0004652457372116349, "loss": 3.1131, "theoretical_loss": 4.217236440383724, "tokens_seen": 260112384 }, { "epoch": 0.08, "learning_rate": 0.0004652357071213641, "loss": 3.1678, "theoretical_loss": 4.217108888603337, "tokens_seen": 260177920 }, { "epoch": 0.08, "learning_rate": 0.0004652256770310933, "loss": 3.1065, "theoretical_loss": 4.21698137794133, "tokens_seen": 260243456 }, { "epoch": 0.08, "learning_rate": 0.00046521564694082247, "loss": 2.9596, "theoretical_loss": 4.216853908374097, "tokens_seen": 260308992 }, { "epoch": 0.08, "learning_rate": 0.0004652056168505517, "loss": 3.008, "theoretical_loss": 4.216726479878052, "tokens_seen": 260374528 }, { "epoch": 0.08, "learning_rate": 0.00046519558676028083, "loss": 3.0454, "theoretical_loss": 4.216599092429631, "tokens_seen": 260440064 }, { "epoch": 0.08, "objective/train/docs_used": 159098, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7527639865875244, "objective/train/theoretical_loss": 4.216471746005286, "objective/train/tokens_used": 280965600, "theoretical_loss": 4.216471746005286, "tokens_seen": 260505600 }, { "epoch": 0.08, "learning_rate": 0.00046518555667001007, "loss": 3.0212, "theoretical_loss": 4.216471746005286, "tokens_seen": 260505600 }, { "epoch": 0.08, "learning_rate": 0.0004651755265797392, "loss": 3.0566, "theoretical_loss": 4.216344440581491, "tokens_seen": 260571136 }, { "epoch": 0.08, "learning_rate": 0.00046516549648946843, "loss": 2.9021, "theoretical_loss": 4.2162171761347365, "tokens_seen": 260636672 }, { "epoch": 0.08, "learning_rate": 0.0004651554663991976, "loss": 2.9991, "theoretical_loss": 4.2160899526415365, "tokens_seen": 260702208 }, { "epoch": 0.08, "learning_rate": 0.0004651454363089268, "loss": 3.0791, "theoretical_loss": 4.215962770078422, "tokens_seen": 260767744 }, { "epoch": 0.08, "learning_rate": 0.000465135406218656, "loss": 3.1413, "theoretical_loss": 4.215835628421942, "tokens_seen": 260833280 }, { "epoch": 0.08, "learning_rate": 0.00046512537612838515, "loss": 2.9224, "theoretical_loss": 4.215708527648667, "tokens_seen": 260898816 }, { "epoch": 0.08, "learning_rate": 0.00046511534603811434, "loss": 3.189, "theoretical_loss": 4.215581467735187, "tokens_seen": 260964352 }, { "epoch": 0.08, "learning_rate": 0.00046510531594784357, "loss": 3.3999, "theoretical_loss": 4.215454448658109, "tokens_seen": 261029888 }, { "epoch": 0.08, "learning_rate": 0.0004650952858575727, "loss": 2.9878, "theoretical_loss": 4.215327470394062, "tokens_seen": 261095424 }, { "epoch": 0.08, "learning_rate": 0.00046508525576730193, "loss": 3.3008, "theoretical_loss": 4.215200532919691, "tokens_seen": 261160960 }, { "epoch": 0.08, "learning_rate": 0.0004650752256770311, "loss": 3.126, "theoretical_loss": 4.215073636211664, "tokens_seen": 261226496 }, { "epoch": 0.08, "learning_rate": 0.0004650651955867603, "loss": 3.0601, "theoretical_loss": 4.214946780246666, "tokens_seen": 261292032 }, { "epoch": 0.08, "learning_rate": 0.0004650551654964895, "loss": 3.2414, "theoretical_loss": 4.214819965001401, "tokens_seen": 261357568 }, { "epoch": 0.08, "learning_rate": 0.00046504513540621866, "loss": 3.3807, "theoretical_loss": 4.214693190452593, "tokens_seen": 261423104 }, { "epoch": 0.08, "learning_rate": 0.00046503510531594784, "loss": 3.2496, "theoretical_loss": 4.214566456576984, "tokens_seen": 261488640 }, { "epoch": 0.08, "learning_rate": 0.0004650250752256771, "loss": 3.0059, "theoretical_loss": 4.214439763351336, "tokens_seen": 261554176 }, { "epoch": 0.08, "learning_rate": 0.0004650150451354062, "loss": 3.3606, "theoretical_loss": 4.214313110752431, "tokens_seen": 261619712 }, { "epoch": 0.08, "learning_rate": 0.00046500501504513544, "loss": 3.2016, "theoretical_loss": 4.214186498757069, "tokens_seen": 261685248 }, { "epoch": 0.08, "learning_rate": 0.00046499498495486456, "loss": 3.238, "theoretical_loss": 4.214059927342068, "tokens_seen": 261750784 }, { "epoch": 0.08, "learning_rate": 0.0004649849548645938, "loss": 3.0229, "theoretical_loss": 4.213933396484267, "tokens_seen": 261816320 }, { "epoch": 0.08, "learning_rate": 0.000464974924774323, "loss": 3.3204, "theoretical_loss": 4.213806906160523, "tokens_seen": 261881856 }, { "epoch": 0.08, "learning_rate": 0.00046496489468405216, "loss": 3.1548, "theoretical_loss": 4.213680456347712, "tokens_seen": 261947392 }, { "epoch": 0.08, "learning_rate": 0.00046495486459378134, "loss": 3.0314, "theoretical_loss": 4.213554047022729, "tokens_seen": 262012928 }, { "epoch": 0.08, "learning_rate": 0.0004649448345035105, "loss": 3.1052, "theoretical_loss": 4.213427678162489, "tokens_seen": 262078464 }, { "epoch": 0.08, "objective/train/docs_used": 160106, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2078521251678467, "objective/train/theoretical_loss": 4.213301349743924, "objective/train/tokens_used": 282604000, "theoretical_loss": 4.213301349743924, "tokens_seen": 262144000 }, { "epoch": 0.08, "learning_rate": 0.0004649348044132397, "loss": 3.285, "theoretical_loss": 4.213301349743924, "tokens_seen": 262144000 }, { "epoch": 0.08, "learning_rate": 0.00046492477432296894, "loss": 3.2845, "theoretical_loss": 4.2131750617439865, "tokens_seen": 262209536 }, { "epoch": 0.08, "learning_rate": 0.00046491474423269807, "loss": 3.0031, "theoretical_loss": 4.213048814139647, "tokens_seen": 262275072 }, { "epoch": 0.08, "learning_rate": 0.0004649047141424273, "loss": 3.0222, "theoretical_loss": 4.212922606907895, "tokens_seen": 262340608 }, { "epoch": 0.08, "learning_rate": 0.0004648946840521565, "loss": 3.0202, "theoretical_loss": 4.21279644002574, "tokens_seen": 262406144 }, { "epoch": 0.08, "learning_rate": 0.00046488465396188566, "loss": 2.9464, "theoretical_loss": 4.212670313470209, "tokens_seen": 262471680 }, { "epoch": 0.08, "learning_rate": 0.00046487462387161485, "loss": 3.3016, "theoretical_loss": 4.212544227218347, "tokens_seen": 262537216 }, { "epoch": 0.08, "learning_rate": 0.000464864593781344, "loss": 3.0782, "theoretical_loss": 4.21241818124722, "tokens_seen": 262602752 }, { "epoch": 0.08, "learning_rate": 0.0004648545636910732, "loss": 3.1333, "theoretical_loss": 4.212292175533912, "tokens_seen": 262668288 }, { "epoch": 0.08, "learning_rate": 0.00046484453360080244, "loss": 2.8875, "theoretical_loss": 4.212166210055526, "tokens_seen": 262733824 }, { "epoch": 0.08, "learning_rate": 0.00046483450351053157, "loss": 2.84, "theoretical_loss": 4.212040284789181, "tokens_seen": 262799360 }, { "epoch": 0.08, "learning_rate": 0.0004648244734202608, "loss": 3.2869, "theoretical_loss": 4.211914399712019, "tokens_seen": 262864896 }, { "epoch": 0.08, "learning_rate": 0.00046481444332998993, "loss": 3.0755, "theoretical_loss": 4.211788554801198, "tokens_seen": 262930432 }, { "epoch": 0.08, "learning_rate": 0.00046480441323971917, "loss": 3.207, "theoretical_loss": 4.211662750033895, "tokens_seen": 262995968 }, { "epoch": 0.08, "learning_rate": 0.00046479438314944835, "loss": 2.9361, "theoretical_loss": 4.211536985387307, "tokens_seen": 263061504 }, { "epoch": 0.08, "learning_rate": 0.00046478435305917753, "loss": 3.1135, "theoretical_loss": 4.211411260838647, "tokens_seen": 263127040 }, { "epoch": 0.08, "learning_rate": 0.0004647743229689067, "loss": 3.003, "theoretical_loss": 4.2112855763651496, "tokens_seen": 263192576 }, { "epoch": 0.08, "learning_rate": 0.00046476429287863595, "loss": 3.252, "theoretical_loss": 4.211159931944065, "tokens_seen": 263258112 }, { "epoch": 0.08, "learning_rate": 0.0004647542627883651, "loss": 3.0456, "theoretical_loss": 4.211034327552666, "tokens_seen": 263323648 }, { "epoch": 0.08, "learning_rate": 0.0004647442326980943, "loss": 3.2239, "theoretical_loss": 4.210908763168239, "tokens_seen": 263389184 }, { "epoch": 0.08, "learning_rate": 0.00046473420260782344, "loss": 3.3088, "theoretical_loss": 4.210783238768093, "tokens_seen": 263454720 }, { "epoch": 0.08, "learning_rate": 0.00046472417251755267, "loss": 3.1186, "theoretical_loss": 4.210657754329553, "tokens_seen": 263520256 }, { "epoch": 0.08, "learning_rate": 0.0004647141424272819, "loss": 3.2965, "theoretical_loss": 4.210532309829965, "tokens_seen": 263585792 }, { "epoch": 0.08, "learning_rate": 0.00046470411233701103, "loss": 3.2105, "theoretical_loss": 4.21040690524669, "tokens_seen": 263651328 }, { "epoch": 0.08, "learning_rate": 0.00046469408224674027, "loss": 2.8521, "theoretical_loss": 4.21028154055711, "tokens_seen": 263716864 }, { "epoch": 0.08, "objective/train/docs_used": 160649, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5538294315338135, "objective/train/theoretical_loss": 4.2101562157386265, "objective/train/tokens_used": 284242400, "theoretical_loss": 4.2101562157386265, "tokens_seen": 263782400 }, { "epoch": 0.08, "learning_rate": 0.0004646840521564694, "loss": 2.7667, "theoretical_loss": 4.2101562157386265, "tokens_seen": 263782400 }, { "epoch": 0.08, "learning_rate": 0.00046467402206619863, "loss": 2.8686, "theoretical_loss": 4.210030930768655, "tokens_seen": 263847936 }, { "epoch": 0.08, "learning_rate": 0.0004646639919759278, "loss": 3.1047, "theoretical_loss": 4.2099056856246335, "tokens_seen": 263913472 }, { "epoch": 0.08, "learning_rate": 0.000464653961885657, "loss": 3.2001, "theoretical_loss": 4.209780480284017, "tokens_seen": 263979008 }, { "epoch": 0.08, "learning_rate": 0.0004646439317953862, "loss": 3.1007, "theoretical_loss": 4.209655314724279, "tokens_seen": 264044544 }, { "epoch": 0.08, "learning_rate": 0.00046463390170511535, "loss": 2.9901, "theoretical_loss": 4.209530188922911, "tokens_seen": 264110080 }, { "epoch": 0.08, "learning_rate": 0.00046462387161484454, "loss": 2.8877, "theoretical_loss": 4.209405102857422, "tokens_seen": 264175616 }, { "epoch": 0.08, "learning_rate": 0.00046461384152457377, "loss": 3.2396, "theoretical_loss": 4.209280056505342, "tokens_seen": 264241152 }, { "epoch": 0.08, "learning_rate": 0.0004646038114343029, "loss": 2.9324, "theoretical_loss": 4.209155049844217, "tokens_seen": 264306688 }, { "epoch": 0.08, "learning_rate": 0.00046459378134403213, "loss": 2.7333, "theoretical_loss": 4.209030082851612, "tokens_seen": 264372224 }, { "epoch": 0.08, "learning_rate": 0.0004645837512537613, "loss": 2.7765, "theoretical_loss": 4.208905155505109, "tokens_seen": 264437760 }, { "epoch": 0.08, "learning_rate": 0.0004645737211634905, "loss": 3.0924, "theoretical_loss": 4.20878026778231, "tokens_seen": 264503296 }, { "epoch": 0.08, "learning_rate": 0.0004645636910732197, "loss": 3.1207, "theoretical_loss": 4.208655419660834, "tokens_seen": 264568832 }, { "epoch": 0.08, "learning_rate": 0.00046455366098294886, "loss": 3.1763, "theoretical_loss": 4.208530611118321, "tokens_seen": 264634368 }, { "epoch": 0.08, "learning_rate": 0.00046454363089267804, "loss": 3.0411, "theoretical_loss": 4.208405842132423, "tokens_seen": 264699904 }, { "epoch": 0.08, "learning_rate": 0.0004645336008024073, "loss": 3.1361, "theoretical_loss": 4.208281112680817, "tokens_seen": 264765440 }, { "epoch": 0.08, "learning_rate": 0.0004645235707121364, "loss": 3.1252, "theoretical_loss": 4.208156422741195, "tokens_seen": 264830976 }, { "epoch": 0.08, "learning_rate": 0.00046451354062186564, "loss": 3.0564, "theoretical_loss": 4.208031772291265, "tokens_seen": 264896512 }, { "epoch": 0.08, "learning_rate": 0.00046450351053159476, "loss": 3.4202, "theoretical_loss": 4.207907161308757, "tokens_seen": 264962048 }, { "epoch": 0.08, "learning_rate": 0.000464493480441324, "loss": 3.3518, "theoretical_loss": 4.2077825897714165, "tokens_seen": 265027584 }, { "epoch": 0.08, "learning_rate": 0.0004644834503510532, "loss": 2.9007, "theoretical_loss": 4.207658057657008, "tokens_seen": 265093120 }, { "epoch": 0.08, "learning_rate": 0.00046447342026078236, "loss": 3.2147, "theoretical_loss": 4.207533564943316, "tokens_seen": 265158656 }, { "epoch": 0.08, "learning_rate": 0.00046446339017051154, "loss": 3.1103, "theoretical_loss": 4.207409111608138, "tokens_seen": 265224192 }, { "epoch": 0.08, "learning_rate": 0.0004644533600802407, "loss": 3.1886, "theoretical_loss": 4.2072846976292935, "tokens_seen": 265289728 }, { "epoch": 0.08, "learning_rate": 0.0004644433299899699, "loss": 3.018, "theoretical_loss": 4.2071603229846195, "tokens_seen": 265355264 }, { "epoch": 0.08, "objective/train/docs_used": 161970, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.322939157485962, "objective/train/theoretical_loss": 4.20703598765197, "objective/train/tokens_used": 285880800, "theoretical_loss": 4.20703598765197, "tokens_seen": 265420800 }, { "epoch": 0.08, "learning_rate": 0.00046443329989969914, "loss": 3.2719, "theoretical_loss": 4.20703598765197, "tokens_seen": 265420800 }, { "epoch": 0.08, "learning_rate": 0.00046442326980942827, "loss": 3.2149, "theoretical_loss": 4.206911691609217, "tokens_seen": 265486336 }, { "epoch": 0.08, "learning_rate": 0.0004644132397191575, "loss": 3.2664, "theoretical_loss": 4.206787434834251, "tokens_seen": 265551872 }, { "epoch": 0.08, "learning_rate": 0.0004644032096288867, "loss": 2.9888, "theoretical_loss": 4.20666321730498, "tokens_seen": 265617408 }, { "epoch": 0.08, "learning_rate": 0.00046439317953861586, "loss": 3.0063, "theoretical_loss": 4.206539038999329, "tokens_seen": 265682944 }, { "epoch": 0.08, "learning_rate": 0.00046438314944834505, "loss": 3.1335, "theoretical_loss": 4.206414899895244, "tokens_seen": 265748480 }, { "epoch": 0.08, "learning_rate": 0.0004643731193580742, "loss": 2.8322, "theoretical_loss": 4.206290799970685, "tokens_seen": 265814016 }, { "epoch": 0.08, "learning_rate": 0.0004643630892678034, "loss": 3.3554, "theoretical_loss": 4.206166739203632, "tokens_seen": 265879552 }, { "epoch": 0.08, "learning_rate": 0.00046435305917753264, "loss": 2.9797, "theoretical_loss": 4.206042717572082, "tokens_seen": 265945088 }, { "epoch": 0.08, "learning_rate": 0.00046434302908726177, "loss": 3.1981, "theoretical_loss": 4.20591873505405, "tokens_seen": 266010624 }, { "epoch": 0.08, "learning_rate": 0.000464332998996991, "loss": 3.1016, "theoretical_loss": 4.20579479162757, "tokens_seen": 266076160 }, { "epoch": 0.08, "learning_rate": 0.00046432296890672013, "loss": 3.2206, "theoretical_loss": 4.205670887270691, "tokens_seen": 266141696 }, { "epoch": 0.08, "learning_rate": 0.00046431293881644937, "loss": 2.916, "theoretical_loss": 4.205547021961482, "tokens_seen": 266207232 }, { "epoch": 0.08, "learning_rate": 0.00046430290872617855, "loss": 3.171, "theoretical_loss": 4.205423195678029, "tokens_seen": 266272768 }, { "epoch": 0.08, "learning_rate": 0.00046429287863590773, "loss": 3.2186, "theoretical_loss": 4.205299408398435, "tokens_seen": 266338304 }, { "epoch": 0.08, "learning_rate": 0.0004642828485456369, "loss": 2.7666, "theoretical_loss": 4.2051756601008226, "tokens_seen": 266403840 }, { "epoch": 0.08, "learning_rate": 0.00046427281845536615, "loss": 2.9658, "theoretical_loss": 4.20505195076333, "tokens_seen": 266469376 }, { "epoch": 0.08, "learning_rate": 0.0004642627883650953, "loss": 3.3264, "theoretical_loss": 4.204928280364115, "tokens_seen": 266534912 }, { "epoch": 0.08, "learning_rate": 0.0004642527582748245, "loss": 3.4252, "theoretical_loss": 4.20480464888135, "tokens_seen": 266600448 }, { "epoch": 0.08, "learning_rate": 0.00046424272818455364, "loss": 2.9859, "theoretical_loss": 4.204681056293228, "tokens_seen": 266665984 }, { "epoch": 0.08, "learning_rate": 0.00046423269809428287, "loss": 3.0057, "theoretical_loss": 4.204557502577957, "tokens_seen": 266731520 }, { "epoch": 0.08, "learning_rate": 0.00046422266800401205, "loss": 3.0726, "theoretical_loss": 4.204433987713767, "tokens_seen": 266797056 }, { "epoch": 0.08, "learning_rate": 0.00046421263791374123, "loss": 3.2009, "theoretical_loss": 4.2043105116789, "tokens_seen": 266862592 }, { "epoch": 0.08, "learning_rate": 0.0004642026078234704, "loss": 2.9325, "theoretical_loss": 4.204187074451617, "tokens_seen": 266928128 }, { "epoch": 0.08, "learning_rate": 0.0004641925777331996, "loss": 3.144, "theoretical_loss": 4.204063676010202, "tokens_seen": 266993664 }, { "epoch": 0.08, "objective/train/docs_used": 162606, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0582354068756104, "objective/train/theoretical_loss": 4.203940316332948, "objective/train/tokens_used": 287519200, "theoretical_loss": 4.203940316332948, "tokens_seen": 267059200 }, { "epoch": 0.08, "learning_rate": 0.0004641825476429288, "loss": 3.183, "theoretical_loss": 4.203940316332948, "tokens_seen": 267059200 }, { "epoch": 0.08, "learning_rate": 0.000464172517552658, "loss": 3.2655, "theoretical_loss": 4.203816995398171, "tokens_seen": 267124736 }, { "epoch": 0.08, "learning_rate": 0.00046416248746238714, "loss": 3.1179, "theoretical_loss": 4.203693713184203, "tokens_seen": 267190272 }, { "epoch": 0.08, "learning_rate": 0.0004641524573721164, "loss": 3.0399, "theoretical_loss": 4.203570469669392, "tokens_seen": 267255808 }, { "epoch": 0.08, "learning_rate": 0.0004641424272818455, "loss": 2.9932, "theoretical_loss": 4.203447264832107, "tokens_seen": 267321344 }, { "epoch": 0.08, "learning_rate": 0.00046413239719157474, "loss": 3.1497, "theoretical_loss": 4.203324098650731, "tokens_seen": 267386880 }, { "epoch": 0.08, "learning_rate": 0.0004641223671013039, "loss": 3.0141, "theoretical_loss": 4.203200971103666, "tokens_seen": 267452416 }, { "epoch": 0.08, "learning_rate": 0.0004641123370110331, "loss": 3.1592, "theoretical_loss": 4.20307788216933, "tokens_seen": 267517952 }, { "epoch": 0.08, "learning_rate": 0.0004641023069207623, "loss": 3.0264, "theoretical_loss": 4.202954831826159, "tokens_seen": 267583488 }, { "epoch": 0.08, "learning_rate": 0.0004640922768304915, "loss": 3.1223, "theoretical_loss": 4.202831820052609, "tokens_seen": 267649024 }, { "epoch": 0.08, "learning_rate": 0.00046408224674022064, "loss": 3.1909, "theoretical_loss": 4.202708846827148, "tokens_seen": 267714560 }, { "epoch": 0.08, "learning_rate": 0.0004640722166499499, "loss": 3.2318, "theoretical_loss": 4.202585912128266, "tokens_seen": 267780096 }, { "epoch": 0.08, "learning_rate": 0.000464062186559679, "loss": 2.8608, "theoretical_loss": 4.202463015934468, "tokens_seen": 267845632 }, { "epoch": 0.08, "learning_rate": 0.00046405215646940824, "loss": 3.2008, "theoretical_loss": 4.202340158224277, "tokens_seen": 267911168 }, { "epoch": 0.08, "learning_rate": 0.0004640421263791374, "loss": 3.1812, "theoretical_loss": 4.202217338976231, "tokens_seen": 267976704 }, { "epoch": 0.08, "learning_rate": 0.0004640320962888666, "loss": 2.8798, "theoretical_loss": 4.2020945581688895, "tokens_seen": 268042240 }, { "epoch": 0.08, "learning_rate": 0.0004640220661985958, "loss": 3.1127, "theoretical_loss": 4.201971815780826, "tokens_seen": 268107776 }, { "epoch": 0.08, "learning_rate": 0.00046401203610832496, "loss": 3.0995, "theoretical_loss": 4.201849111790631, "tokens_seen": 268173312 }, { "epoch": 0.08, "learning_rate": 0.00046400200601805414, "loss": 3.0434, "theoretical_loss": 4.201726446176915, "tokens_seen": 268238848 }, { "epoch": 0.08, "learning_rate": 0.0004639919759277834, "loss": 3.05, "theoretical_loss": 4.201603818918302, "tokens_seen": 268304384 }, { "epoch": 0.08, "learning_rate": 0.0004639819458375125, "loss": 3.2322, "theoretical_loss": 4.201481229993435, "tokens_seen": 268369920 }, { "epoch": 0.08, "learning_rate": 0.00046397191574724174, "loss": 3.2241, "theoretical_loss": 4.201358679380976, "tokens_seen": 268435456 }, { "epoch": 0.08, "learning_rate": 0.0004639618856569709, "loss": 3.1414, "theoretical_loss": 4.201236167059601, "tokens_seen": 268500992 }, { "epoch": 0.08, "learning_rate": 0.0004639518555667001, "loss": 3.2403, "theoretical_loss": 4.201113693008002, "tokens_seen": 268566528 }, { "epoch": 0.08, "learning_rate": 0.00046394182547642934, "loss": 3.0643, "theoretical_loss": 4.200991257204894, "tokens_seen": 268632064 }, { "epoch": 0.08, "objective/train/docs_used": 163909, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.577915668487549, "objective/train/theoretical_loss": 4.2008688596290025, "objective/train/tokens_used": 289157600, "theoretical_loss": 4.2008688596290025, "tokens_seen": 268697600 }, { "epoch": 0.08, "learning_rate": 0.00046393179538615847, "loss": 2.8287, "theoretical_loss": 4.2008688596290025, "tokens_seen": 268697600 }, { "epoch": 0.08, "learning_rate": 0.0004639217652958877, "loss": 3.1737, "theoretical_loss": 4.200746500259073, "tokens_seen": 268763136 }, { "epoch": 0.08, "learning_rate": 0.0004639117352056169, "loss": 2.6194, "theoretical_loss": 4.200624179073869, "tokens_seen": 268828672 }, { "epoch": 0.08, "learning_rate": 0.00046390170511534606, "loss": 3.2024, "theoretical_loss": 4.2005018960521685, "tokens_seen": 268894208 }, { "epoch": 0.08, "learning_rate": 0.00046389167502507525, "loss": 3.3453, "theoretical_loss": 4.200379651172769, "tokens_seen": 268959744 }, { "epoch": 0.08, "learning_rate": 0.0004638816449348044, "loss": 2.8859, "theoretical_loss": 4.200257444414483, "tokens_seen": 269025280 }, { "epoch": 0.08, "learning_rate": 0.0004638716148445336, "loss": 3.1556, "theoretical_loss": 4.200135275756139, "tokens_seen": 269090816 }, { "epoch": 0.08, "learning_rate": 0.00046386158475426284, "loss": 3.0459, "theoretical_loss": 4.200013145176587, "tokens_seen": 269156352 }, { "epoch": 0.08, "learning_rate": 0.00046385155466399197, "loss": 3.088, "theoretical_loss": 4.199891052654689, "tokens_seen": 269221888 }, { "epoch": 0.08, "learning_rate": 0.0004638415245737212, "loss": 3.0612, "theoretical_loss": 4.199768998169326, "tokens_seen": 269287424 }, { "epoch": 0.08, "learning_rate": 0.00046383149448345033, "loss": 3.0789, "theoretical_loss": 4.199646981699395, "tokens_seen": 269352960 }, { "epoch": 0.08, "learning_rate": 0.00046382146439317957, "loss": 3.1749, "theoretical_loss": 4.199525003223812, "tokens_seen": 269418496 }, { "epoch": 0.08, "learning_rate": 0.00046381143430290875, "loss": 3.162, "theoretical_loss": 4.199403062721506, "tokens_seen": 269484032 }, { "epoch": 0.08, "learning_rate": 0.00046380140421263793, "loss": 3.1445, "theoretical_loss": 4.199281160171427, "tokens_seen": 269549568 }, { "epoch": 0.08, "learning_rate": 0.0004637913741223671, "loss": 3.0702, "theoretical_loss": 4.1991592955525405, "tokens_seen": 269615104 }, { "epoch": 0.08, "learning_rate": 0.00046378134403209635, "loss": 3.0764, "theoretical_loss": 4.199037468843825, "tokens_seen": 269680640 }, { "epoch": 0.08, "learning_rate": 0.0004637713139418255, "loss": 3.1614, "theoretical_loss": 4.198915680024282, "tokens_seen": 269746176 }, { "epoch": 0.08, "learning_rate": 0.0004637612838515547, "loss": 2.84, "theoretical_loss": 4.198793929072925, "tokens_seen": 269811712 }, { "epoch": 0.08, "learning_rate": 0.00046375125376128384, "loss": 3.1708, "theoretical_loss": 4.198672215968785, "tokens_seen": 269877248 }, { "epoch": 0.08, "learning_rate": 0.00046374122367101307, "loss": 3.0235, "theoretical_loss": 4.198550540690912, "tokens_seen": 269942784 }, { "epoch": 0.08, "learning_rate": 0.00046373119358074225, "loss": 3.393, "theoretical_loss": 4.198428903218371, "tokens_seen": 270008320 }, { "epoch": 0.08, "learning_rate": 0.00046372116349047143, "loss": 3.1363, "theoretical_loss": 4.198307303530243, "tokens_seen": 270073856 }, { "epoch": 0.08, "learning_rate": 0.0004637111334002006, "loss": 3.0941, "theoretical_loss": 4.198185741605628, "tokens_seen": 270139392 }, { "epoch": 0.08, "learning_rate": 0.0004637011033099298, "loss": 3.0913, "theoretical_loss": 4.19806421742364, "tokens_seen": 270204928 }, { "epoch": 0.08, "learning_rate": 0.000463691073219659, "loss": 3.2218, "theoretical_loss": 4.197942730963412, "tokens_seen": 270270464 }, { "epoch": 0.08, "objective/train/docs_used": 164570, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8737075328826904, "objective/train/theoretical_loss": 4.19782128220409, "objective/train/tokens_used": 290796000, "theoretical_loss": 4.19782128220409, "tokens_seen": 270336000 }, { "epoch": 0.08, "learning_rate": 0.0004636810431293882, "loss": 2.9242, "theoretical_loss": 4.19782128220409, "tokens_seen": 270336000 }, { "epoch": 0.08, "learning_rate": 0.00046367101303911734, "loss": 2.8725, "theoretical_loss": 4.19769987112484, "tokens_seen": 270401536 }, { "epoch": 0.08, "learning_rate": 0.0004636609829488466, "loss": 3.3748, "theoretical_loss": 4.1975784977048445, "tokens_seen": 270467072 }, { "epoch": 0.08, "learning_rate": 0.0004636509528585757, "loss": 3.0672, "theoretical_loss": 4.1974571619233, "tokens_seen": 270532608 }, { "epoch": 0.08, "learning_rate": 0.00046364092276830494, "loss": 3.1778, "theoretical_loss": 4.197335863759422, "tokens_seen": 270598144 }, { "epoch": 0.08, "learning_rate": 0.0004636308926780341, "loss": 3.2507, "theoretical_loss": 4.1972146031924416, "tokens_seen": 270663680 }, { "epoch": 0.08, "learning_rate": 0.0004636208625877633, "loss": 2.9583, "theoretical_loss": 4.197093380201606, "tokens_seen": 270729216 }, { "epoch": 0.08, "learning_rate": 0.0004636108324974925, "loss": 3.1339, "theoretical_loss": 4.196972194766179, "tokens_seen": 270794752 }, { "epoch": 0.08, "learning_rate": 0.0004636008024072217, "loss": 3.2131, "theoretical_loss": 4.196851046865442, "tokens_seen": 270860288 }, { "epoch": 0.08, "learning_rate": 0.00046359077231695084, "loss": 3.0693, "theoretical_loss": 4.1967299364786905, "tokens_seen": 270925824 }, { "epoch": 0.08, "learning_rate": 0.0004635807422266801, "loss": 3.1201, "theoretical_loss": 4.196608863585239, "tokens_seen": 270991360 }, { "epoch": 0.08, "learning_rate": 0.0004635707121364092, "loss": 3.0753, "theoretical_loss": 4.1964878281644165, "tokens_seen": 271056896 }, { "epoch": 0.08, "learning_rate": 0.00046356068204613844, "loss": 3.1767, "theoretical_loss": 4.19636683019557, "tokens_seen": 271122432 }, { "epoch": 0.08, "learning_rate": 0.0004635506519558676, "loss": 3.1568, "theoretical_loss": 4.196245869658061, "tokens_seen": 271187968 }, { "epoch": 0.08, "learning_rate": 0.0004635406218655968, "loss": 3.1067, "theoretical_loss": 4.1961249465312696, "tokens_seen": 271253504 }, { "epoch": 0.08, "learning_rate": 0.000463530591775326, "loss": 3.1049, "theoretical_loss": 4.196004060794589, "tokens_seen": 271319040 }, { "epoch": 0.08, "learning_rate": 0.00046352056168505516, "loss": 3.1558, "theoretical_loss": 4.195883212427433, "tokens_seen": 271384576 }, { "epoch": 0.08, "learning_rate": 0.00046351053159478434, "loss": 3.104, "theoretical_loss": 4.195762401409229, "tokens_seen": 271450112 }, { "epoch": 0.08, "learning_rate": 0.0004635005015045136, "loss": 2.9589, "theoretical_loss": 4.19564162771942, "tokens_seen": 271515648 }, { "epoch": 0.08, "learning_rate": 0.0004634904714142427, "loss": 3.0468, "theoretical_loss": 4.195520891337466, "tokens_seen": 271581184 }, { "epoch": 0.08, "learning_rate": 0.00046348044132397194, "loss": 2.9809, "theoretical_loss": 4.195400192242845, "tokens_seen": 271646720 }, { "epoch": 0.08, "learning_rate": 0.00046347041123370107, "loss": 3.2978, "theoretical_loss": 4.19527953041505, "tokens_seen": 271712256 }, { "epoch": 0.08, "learning_rate": 0.0004634603811434303, "loss": 3.4054, "theoretical_loss": 4.19515890583359, "tokens_seen": 271777792 }, { "epoch": 0.08, "learning_rate": 0.0004634503510531595, "loss": 3.1847, "theoretical_loss": 4.195038318477989, "tokens_seen": 271843328 }, { "epoch": 0.08, "learning_rate": 0.00046344032096288867, "loss": 3.1394, "theoretical_loss": 4.194917768327789, "tokens_seen": 271908864 }, { "epoch": 0.08, "objective/train/docs_used": 165961, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7331738471984863, "objective/train/theoretical_loss": 4.194797255362549, "objective/train/tokens_used": 292434400, "theoretical_loss": 4.194797255362549, "tokens_seen": 271974400 }, { "epoch": 0.08, "learning_rate": 0.00046343029087261785, "loss": 2.8674, "theoretical_loss": 4.194797255362549, "tokens_seen": 271974400 }, { "epoch": 0.08, "learning_rate": 0.0004634202607823471, "loss": 3.05, "theoretical_loss": 4.194676779561841, "tokens_seen": 272039936 }, { "epoch": 0.08, "learning_rate": 0.0004634102306920762, "loss": 3.0387, "theoretical_loss": 4.194556340905256, "tokens_seen": 272105472 }, { "epoch": 0.08, "learning_rate": 0.00046340020060180545, "loss": 3.0849, "theoretical_loss": 4.194435939372401, "tokens_seen": 272171008 }, { "epoch": 0.08, "learning_rate": 0.00046339017051153457, "loss": 3.1541, "theoretical_loss": 4.194315574942896, "tokens_seen": 272236544 }, { "epoch": 0.08, "learning_rate": 0.0004633801404212638, "loss": 3.2083, "theoretical_loss": 4.194195247596381, "tokens_seen": 272302080 }, { "epoch": 0.08, "learning_rate": 0.000463370110330993, "loss": 3.2071, "theoretical_loss": 4.19407495731251, "tokens_seen": 272367616 }, { "epoch": 0.08, "learning_rate": 0.00046336008024072217, "loss": 3.2115, "theoretical_loss": 4.193954704070952, "tokens_seen": 272433152 }, { "epoch": 0.08, "learning_rate": 0.00046335005015045135, "loss": 3.1099, "theoretical_loss": 4.193834487851396, "tokens_seen": 272498688 }, { "epoch": 0.08, "learning_rate": 0.00046334002006018053, "loss": 2.9455, "theoretical_loss": 4.193714308633542, "tokens_seen": 272564224 }, { "epoch": 0.08, "learning_rate": 0.0004633299899699097, "loss": 3.2173, "theoretical_loss": 4.1935941663971095, "tokens_seen": 272629760 }, { "epoch": 0.08, "learning_rate": 0.00046331995987963895, "loss": 3.0527, "theoretical_loss": 4.193474061121833, "tokens_seen": 272695296 }, { "epoch": 0.08, "learning_rate": 0.0004633099297893681, "loss": 3.0777, "theoretical_loss": 4.193353992787463, "tokens_seen": 272760832 }, { "epoch": 0.08, "learning_rate": 0.0004632998996990973, "loss": 3.1036, "theoretical_loss": 4.193233961373766, "tokens_seen": 272826368 }, { "epoch": 0.08, "learning_rate": 0.00046328986960882644, "loss": 2.9951, "theoretical_loss": 4.1931139668605235, "tokens_seen": 272891904 }, { "epoch": 0.08, "learning_rate": 0.0004632798395185557, "loss": 3.0116, "theoretical_loss": 4.192994009227535, "tokens_seen": 272957440 }, { "epoch": 0.08, "learning_rate": 0.00046326980942828485, "loss": 2.8518, "theoretical_loss": 4.192874088454613, "tokens_seen": 273022976 }, { "epoch": 0.08, "learning_rate": 0.00046325977933801404, "loss": 2.8654, "theoretical_loss": 4.19275420452159, "tokens_seen": 273088512 }, { "epoch": 0.08, "learning_rate": 0.0004632497492477432, "loss": 3.0445, "theoretical_loss": 4.192634357408309, "tokens_seen": 273154048 }, { "epoch": 0.08, "learning_rate": 0.00046323971915747245, "loss": 2.9804, "theoretical_loss": 4.192514547094634, "tokens_seen": 273219584 }, { "epoch": 0.08, "learning_rate": 0.0004632296890672016, "loss": 3.0997, "theoretical_loss": 4.192394773560441, "tokens_seen": 273285120 }, { "epoch": 0.08, "learning_rate": 0.0004632196589769308, "loss": 3.0125, "theoretical_loss": 4.192275036785625, "tokens_seen": 273350656 }, { "epoch": 0.08, "learning_rate": 0.00046320962888666, "loss": 2.8637, "theoretical_loss": 4.192155336750094, "tokens_seen": 273416192 }, { "epoch": 0.08, "learning_rate": 0.0004631995987963892, "loss": 3.1459, "theoretical_loss": 4.192035673433773, "tokens_seen": 273481728 }, { "epoch": 0.08, "learning_rate": 0.0004631895687061184, "loss": 3.0333, "theoretical_loss": 4.191916046816605, "tokens_seen": 273547264 }, { "epoch": 0.08, "objective/train/docs_used": 166474, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.229163885116577, "objective/train/theoretical_loss": 4.191796456878544, "objective/train/tokens_used": 294072800, "theoretical_loss": 4.191796456878544, "tokens_seen": 273612800 }, { "epoch": 0.08, "learning_rate": 0.00046317953861584754, "loss": 3.3893, "theoretical_loss": 4.191796456878544, "tokens_seen": 273612800 }, { "epoch": 0.08, "learning_rate": 0.0004631695085255768, "loss": 3.2798, "theoretical_loss": 4.191676903599563, "tokens_seen": 273678336 }, { "epoch": 0.08, "learning_rate": 0.0004631594784353059, "loss": 3.2739, "theoretical_loss": 4.191557386959651, "tokens_seen": 273743872 }, { "epoch": 0.08, "learning_rate": 0.00046314944834503514, "loss": 2.8998, "theoretical_loss": 4.191437906938811, "tokens_seen": 273809408 }, { "epoch": 0.08, "learning_rate": 0.0004631394182547643, "loss": 3.1002, "theoretical_loss": 4.191318463517062, "tokens_seen": 273874944 }, { "epoch": 0.08, "learning_rate": 0.0004631293881644935, "loss": 3.052, "theoretical_loss": 4.19119905667444, "tokens_seen": 273940480 }, { "epoch": 0.08, "learning_rate": 0.0004631193580742227, "loss": 3.0649, "theoretical_loss": 4.191079686390996, "tokens_seen": 274006016 }, { "epoch": 0.08, "learning_rate": 0.0004631093279839519, "loss": 3.1153, "theoretical_loss": 4.190960352646796, "tokens_seen": 274071552 }, { "epoch": 0.08, "learning_rate": 0.00046309929789368104, "loss": 3.0196, "theoretical_loss": 4.190841055421921, "tokens_seen": 274137088 }, { "epoch": 0.08, "learning_rate": 0.0004630892678034103, "loss": 3.0416, "theoretical_loss": 4.19072179469647, "tokens_seen": 274202624 }, { "epoch": 0.08, "learning_rate": 0.0004630792377131394, "loss": 3.0144, "theoretical_loss": 4.190602570450556, "tokens_seen": 274268160 }, { "epoch": 0.08, "learning_rate": 0.00046306920762286864, "loss": 3.1422, "theoretical_loss": 4.190483382664308, "tokens_seen": 274333696 }, { "epoch": 0.08, "learning_rate": 0.0004630591775325978, "loss": 3.1745, "theoretical_loss": 4.19036423131787, "tokens_seen": 274399232 }, { "epoch": 0.08, "learning_rate": 0.000463049147442327, "loss": 3.0466, "theoretical_loss": 4.190245116391403, "tokens_seen": 274464768 }, { "epoch": 0.08, "learning_rate": 0.0004630391173520562, "loss": 2.9194, "theoretical_loss": 4.190126037865082, "tokens_seen": 274530304 }, { "epoch": 0.08, "learning_rate": 0.00046302908726178536, "loss": 3.2119, "theoretical_loss": 4.190006995719098, "tokens_seen": 274595840 }, { "epoch": 0.08, "learning_rate": 0.00046301905717151455, "loss": 3.1011, "theoretical_loss": 4.1898879899336565, "tokens_seen": 274661376 }, { "epoch": 0.08, "learning_rate": 0.0004630090270812438, "loss": 3.0468, "theoretical_loss": 4.189769020488981, "tokens_seen": 274726912 }, { "epoch": 0.08, "learning_rate": 0.0004629989969909729, "loss": 3.368, "theoretical_loss": 4.189650087365309, "tokens_seen": 274792448 }, { "epoch": 0.08, "learning_rate": 0.00046298896690070214, "loss": 3.0634, "theoretical_loss": 4.189531190542893, "tokens_seen": 274857984 }, { "epoch": 0.08, "learning_rate": 0.00046297893681043127, "loss": 3.0413, "theoretical_loss": 4.189412330002001, "tokens_seen": 274923520 }, { "epoch": 0.08, "learning_rate": 0.0004629689067201605, "loss": 3.3153, "theoretical_loss": 4.189293505722918, "tokens_seen": 274989056 }, { "epoch": 0.08, "learning_rate": 0.0004629588766298897, "loss": 2.9856, "theoretical_loss": 4.189174717685942, "tokens_seen": 275054592 }, { "epoch": 0.08, "learning_rate": 0.00046294884653961887, "loss": 3.1424, "theoretical_loss": 4.189055965871389, "tokens_seen": 275120128 }, { "epoch": 0.08, "learning_rate": 0.00046293881644934805, "loss": 3.3162, "theoretical_loss": 4.188937250259587, "tokens_seen": 275185664 }, { "epoch": 0.08, "objective/train/docs_used": 167916, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9187827110290527, "objective/train/theoretical_loss": 4.188818570830883, "objective/train/tokens_used": 295711200, "theoretical_loss": 4.188818570830883, "tokens_seen": 275251200 }, { "epoch": 0.08, "learning_rate": 0.0004629287863590773, "loss": 3.233, "theoretical_loss": 4.188818570830883, "tokens_seen": 275251200 }, { "epoch": 0.08, "learning_rate": 0.0004629187562688064, "loss": 3.1334, "theoretical_loss": 4.188699927565638, "tokens_seen": 275316736 }, { "epoch": 0.08, "learning_rate": 0.00046290872617853565, "loss": 3.2221, "theoretical_loss": 4.188581320444228, "tokens_seen": 275382272 }, { "epoch": 0.08, "learning_rate": 0.00046289869608826477, "loss": 3.4486, "theoretical_loss": 4.1884627494470426, "tokens_seen": 275447808 }, { "epoch": 0.08, "learning_rate": 0.000462888665997994, "loss": 2.7931, "theoretical_loss": 4.1883442145544905, "tokens_seen": 275513344 }, { "epoch": 0.08, "learning_rate": 0.0004628786359077232, "loss": 3.002, "theoretical_loss": 4.188225715746992, "tokens_seen": 275578880 }, { "epoch": 0.08, "learning_rate": 0.00046286860581745237, "loss": 3.2486, "theoretical_loss": 4.188107253004986, "tokens_seen": 275644416 }, { "epoch": 0.08, "learning_rate": 0.00046285857572718155, "loss": 3.2086, "theoretical_loss": 4.187988826308925, "tokens_seen": 275709952 }, { "epoch": 0.08, "learning_rate": 0.00046284854563691073, "loss": 2.9635, "theoretical_loss": 4.187870435639275, "tokens_seen": 275775488 }, { "epoch": 0.08, "learning_rate": 0.0004628385155466399, "loss": 2.9808, "theoretical_loss": 4.18775208097652, "tokens_seen": 275841024 }, { "epoch": 0.08, "learning_rate": 0.00046282848545636915, "loss": 3.2022, "theoretical_loss": 4.187633762301159, "tokens_seen": 275906560 }, { "epoch": 0.08, "learning_rate": 0.0004628184553660983, "loss": 3.0849, "theoretical_loss": 4.187515479593704, "tokens_seen": 275972096 }, { "epoch": 0.08, "learning_rate": 0.0004628084252758275, "loss": 3.0014, "theoretical_loss": 4.187397232834683, "tokens_seen": 276037632 }, { "epoch": 0.08, "learning_rate": 0.00046279839518555664, "loss": 3.1524, "theoretical_loss": 4.187279022004642, "tokens_seen": 276103168 }, { "epoch": 0.08, "learning_rate": 0.0004627883650952859, "loss": 3.1932, "theoretical_loss": 4.1871608470841375, "tokens_seen": 276168704 }, { "epoch": 0.08, "learning_rate": 0.00046277833500501505, "loss": 2.89, "theoretical_loss": 4.1870427080537445, "tokens_seen": 276234240 }, { "epoch": 0.08, "learning_rate": 0.00046276830491474424, "loss": 3.2589, "theoretical_loss": 4.1869246048940525, "tokens_seen": 276299776 }, { "epoch": 0.08, "learning_rate": 0.0004627582748244734, "loss": 3.1626, "theoretical_loss": 4.186806537585666, "tokens_seen": 276365312 }, { "epoch": 0.08, "learning_rate": 0.00046274824473420265, "loss": 3.1922, "theoretical_loss": 4.186688506109202, "tokens_seen": 276430848 }, { "epoch": 0.08, "learning_rate": 0.0004627382146439318, "loss": 2.9536, "theoretical_loss": 4.186570510445296, "tokens_seen": 276496384 }, { "epoch": 0.08, "learning_rate": 0.000462728184553661, "loss": 3.3615, "theoretical_loss": 4.186452550574599, "tokens_seen": 276561920 }, { "epoch": 0.08, "learning_rate": 0.00046271815446339014, "loss": 3.2877, "theoretical_loss": 4.186334626477774, "tokens_seen": 276627456 }, { "epoch": 0.08, "learning_rate": 0.0004627081243731194, "loss": 3.2169, "theoretical_loss": 4.186216738135501, "tokens_seen": 276692992 }, { "epoch": 0.08, "learning_rate": 0.00046269809428284856, "loss": 3.0765, "theoretical_loss": 4.186098885528473, "tokens_seen": 276758528 }, { "epoch": 0.08, "learning_rate": 0.00046268806419257774, "loss": 3.1812, "theoretical_loss": 4.185981068637401, "tokens_seen": 276824064 }, { "epoch": 0.08, "objective/train/docs_used": 168545, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.318019390106201, "objective/train/theoretical_loss": 4.185863287443008, "objective/train/tokens_used": 297349600, "theoretical_loss": 4.185863287443008, "tokens_seen": 276889600 }, { "epoch": 0.08, "learning_rate": 0.0004626780341023069, "loss": 3.1149, "theoretical_loss": 4.185863287443008, "tokens_seen": 276889600 }, { "epoch": 0.08, "learning_rate": 0.0004626680040120361, "loss": 3.1434, "theoretical_loss": 4.185745541926035, "tokens_seen": 276955136 }, { "epoch": 0.08, "learning_rate": 0.0004626579739217653, "loss": 3.0085, "theoretical_loss": 4.185627832067237, "tokens_seen": 277020672 }, { "epoch": 0.08, "learning_rate": 0.0004626479438314945, "loss": 3.1581, "theoretical_loss": 4.1855101578473795, "tokens_seen": 277086208 }, { "epoch": 0.08, "learning_rate": 0.00046263791374122364, "loss": 3.2237, "theoretical_loss": 4.18539251924725, "tokens_seen": 277151744 }, { "epoch": 0.08, "learning_rate": 0.0004626278836509529, "loss": 3.0262, "theoretical_loss": 4.185274916247646, "tokens_seen": 277217280 }, { "epoch": 0.08, "learning_rate": 0.000462617853560682, "loss": 3.3536, "theoretical_loss": 4.185157348829383, "tokens_seen": 277282816 }, { "epoch": 0.08, "learning_rate": 0.00046260782347041124, "loss": 3.1312, "theoretical_loss": 4.185039816973289, "tokens_seen": 277348352 }, { "epoch": 0.08, "learning_rate": 0.0004625977933801404, "loss": 3.1212, "theoretical_loss": 4.184922320660207, "tokens_seen": 277413888 }, { "epoch": 0.08, "learning_rate": 0.0004625877632898696, "loss": 3.0527, "theoretical_loss": 4.184804859870997, "tokens_seen": 277479424 }, { "epoch": 0.08, "learning_rate": 0.0004625777331995988, "loss": 3.1812, "theoretical_loss": 4.184687434586531, "tokens_seen": 277544960 }, { "epoch": 0.08, "learning_rate": 0.000462567703109328, "loss": 2.9691, "theoretical_loss": 4.184570044787698, "tokens_seen": 277610496 }, { "epoch": 0.08, "learning_rate": 0.00046255767301905715, "loss": 3.073, "theoretical_loss": 4.1844526904554, "tokens_seen": 277676032 }, { "epoch": 0.08, "learning_rate": 0.0004625476429287864, "loss": 3.0977, "theoretical_loss": 4.184335371570556, "tokens_seen": 277741568 }, { "epoch": 0.08, "learning_rate": 0.0004625376128385155, "loss": 3.0928, "theoretical_loss": 4.184218088114097, "tokens_seen": 277807104 }, { "epoch": 0.08, "learning_rate": 0.00046252758274824475, "loss": 3.1863, "theoretical_loss": 4.1841008400669715, "tokens_seen": 277872640 }, { "epoch": 0.08, "learning_rate": 0.0004625175526579739, "loss": 3.3092, "theoretical_loss": 4.183983627410142, "tokens_seen": 277938176 }, { "epoch": 0.08, "learning_rate": 0.0004625075225677031, "loss": 2.8105, "theoretical_loss": 4.183866450124584, "tokens_seen": 278003712 }, { "epoch": 0.08, "learning_rate": 0.0004624974924774323, "loss": 3.1829, "theoretical_loss": 4.18374930819129, "tokens_seen": 278069248 }, { "epoch": 0.08, "learning_rate": 0.00046248746238716147, "loss": 3.2848, "theoretical_loss": 4.183632201591264, "tokens_seen": 278134784 }, { "epoch": 0.08, "learning_rate": 0.00046247743229689065, "loss": 2.9075, "theoretical_loss": 4.18351513030553, "tokens_seen": 278200320 }, { "epoch": 0.08, "learning_rate": 0.0004624674022066199, "loss": 2.9081, "theoretical_loss": 4.1833980943151206, "tokens_seen": 278265856 }, { "epoch": 0.08, "learning_rate": 0.00046245737211634907, "loss": 3.117, "theoretical_loss": 4.183281093601087, "tokens_seen": 278331392 }, { "epoch": 0.08, "learning_rate": 0.00046244734202607825, "loss": 3.1446, "theoretical_loss": 4.183164128144495, "tokens_seen": 278396928 }, { "epoch": 0.08, "learning_rate": 0.0004624373119358075, "loss": 3.1212, "theoretical_loss": 4.183047197926422, "tokens_seen": 278462464 }, { "epoch": 0.08, "objective/train/docs_used": 169743, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.211092233657837, "objective/train/theoretical_loss": 4.182930302927963, "objective/train/tokens_used": 298988000, "theoretical_loss": 4.182930302927963, "tokens_seen": 278528000 }, { "epoch": 0.08, "learning_rate": 0.0004624272818455366, "loss": 2.9243, "theoretical_loss": 4.182930302927963, "tokens_seen": 278528000 }, { "epoch": 0.08, "learning_rate": 0.00046241725175526585, "loss": 3.1498, "theoretical_loss": 4.182813443130227, "tokens_seen": 278593536 }, { "epoch": 0.08, "learning_rate": 0.000462407221664995, "loss": 3.1976, "theoretical_loss": 4.182696618514337, "tokens_seen": 278659072 }, { "epoch": 0.08, "learning_rate": 0.0004623971915747242, "loss": 2.9944, "theoretical_loss": 4.18257982906143, "tokens_seen": 278724608 }, { "epoch": 0.08, "learning_rate": 0.0004623871614844534, "loss": 3.0975, "theoretical_loss": 4.1824630747526585, "tokens_seen": 278790144 }, { "epoch": 0.08, "learning_rate": 0.00046237713139418257, "loss": 3.1447, "theoretical_loss": 4.182346355569189, "tokens_seen": 278855680 }, { "epoch": 0.08, "learning_rate": 0.00046236710130391175, "loss": 3.0155, "theoretical_loss": 4.182229671492204, "tokens_seen": 278921216 }, { "epoch": 0.08, "learning_rate": 0.00046235707121364093, "loss": 3.1009, "theoretical_loss": 4.1821130225028975, "tokens_seen": 278986752 }, { "epoch": 0.08, "learning_rate": 0.0004623470411233701, "loss": 3.1349, "theoretical_loss": 4.1819964085824815, "tokens_seen": 279052288 }, { "epoch": 0.08, "learning_rate": 0.00046233701103309935, "loss": 3.2392, "theoretical_loss": 4.181879829712178, "tokens_seen": 279117824 }, { "epoch": 0.08, "learning_rate": 0.0004623269809428285, "loss": 2.8839, "theoretical_loss": 4.181763285873231, "tokens_seen": 279183360 }, { "epoch": 0.08, "learning_rate": 0.0004623169508525577, "loss": 3.2192, "theoretical_loss": 4.181646777046889, "tokens_seen": 279248896 }, { "epoch": 0.08, "learning_rate": 0.00046230692076228684, "loss": 3.1266, "theoretical_loss": 4.181530303214423, "tokens_seen": 279314432 }, { "epoch": 0.08, "learning_rate": 0.0004622968906720161, "loss": 3.2385, "theoretical_loss": 4.181413864357115, "tokens_seen": 279379968 }, { "epoch": 0.08, "learning_rate": 0.00046228686058174525, "loss": 3.0699, "theoretical_loss": 4.181297460456262, "tokens_seen": 279445504 }, { "epoch": 0.08, "learning_rate": 0.00046227683049147444, "loss": 3.2547, "theoretical_loss": 4.181181091493174, "tokens_seen": 279511040 }, { "epoch": 0.08, "learning_rate": 0.0004622668004012036, "loss": 3.0276, "theoretical_loss": 4.181064757449178, "tokens_seen": 279576576 }, { "epoch": 0.08, "learning_rate": 0.00046225677031093285, "loss": 3.1922, "theoretical_loss": 4.180948458305615, "tokens_seen": 279642112 }, { "epoch": 0.08, "learning_rate": 0.000462246740220662, "loss": 3.0343, "theoretical_loss": 4.180832194043836, "tokens_seen": 279707648 }, { "epoch": 0.08, "learning_rate": 0.0004622367101303912, "loss": 3.1996, "theoretical_loss": 4.180715964645213, "tokens_seen": 279773184 }, { "epoch": 0.08, "learning_rate": 0.00046222668004012034, "loss": 3.3317, "theoretical_loss": 4.180599770091126, "tokens_seen": 279838720 }, { "epoch": 0.08, "learning_rate": 0.0004622166499498496, "loss": 2.9687, "theoretical_loss": 4.180483610362975, "tokens_seen": 279904256 }, { "epoch": 0.08, "learning_rate": 0.00046220661985957876, "loss": 3.158, "theoretical_loss": 4.18036748544217, "tokens_seen": 279969792 }, { "epoch": 0.08, "learning_rate": 0.00046219658976930794, "loss": 3.0762, "theoretical_loss": 4.180251395310137, "tokens_seen": 280035328 }, { "epoch": 0.08, "learning_rate": 0.0004621865596790371, "loss": 3.1545, "theoretical_loss": 4.1801353399483165, "tokens_seen": 280100864 }, { "epoch": 0.08, "objective/train/docs_used": 170343, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.5833492279052734, "objective/train/theoretical_loss": 4.180019319338163, "objective/train/tokens_used": 300626400, "theoretical_loss": 4.180019319338163, "tokens_seen": 280166400 }, { "epoch": 0.08, "learning_rate": 0.0004621765295887663, "loss": 3.2849, "theoretical_loss": 4.180019319338163, "tokens_seen": 280166400 }, { "epoch": 0.08, "learning_rate": 0.0004621664994984955, "loss": 2.8302, "theoretical_loss": 4.179903333461144, "tokens_seen": 280231936 }, { "epoch": 0.08, "learning_rate": 0.0004621564694082247, "loss": 2.8809, "theoretical_loss": 4.179787382298744, "tokens_seen": 280297472 }, { "epoch": 0.08, "learning_rate": 0.00046214643931795384, "loss": 3.2571, "theoretical_loss": 4.179671465832458, "tokens_seen": 280363008 }, { "epoch": 0.08, "learning_rate": 0.0004621364092276831, "loss": 2.9144, "theoretical_loss": 4.179555584043799, "tokens_seen": 280428544 }, { "epoch": 0.08, "learning_rate": 0.0004621263791374122, "loss": 3.1233, "theoretical_loss": 4.17943973691429, "tokens_seen": 280494080 }, { "epoch": 0.09, "learning_rate": 0.00046211634904714144, "loss": 3.0764, "theoretical_loss": 4.179323924425472, "tokens_seen": 280559616 }, { "epoch": 0.09, "learning_rate": 0.0004621063189568706, "loss": 3.1949, "theoretical_loss": 4.179208146558899, "tokens_seen": 280625152 }, { "epoch": 0.09, "learning_rate": 0.0004620962888665998, "loss": 3.1481, "theoretical_loss": 4.1790924032961385, "tokens_seen": 280690688 }, { "epoch": 0.09, "learning_rate": 0.000462086258776329, "loss": 2.9628, "theoretical_loss": 4.178976694618772, "tokens_seen": 280756224 }, { "epoch": 0.09, "learning_rate": 0.0004620762286860582, "loss": 3.1512, "theoretical_loss": 4.178861020508395, "tokens_seen": 280821760 }, { "epoch": 0.09, "learning_rate": 0.00046206619859578735, "loss": 2.9415, "theoretical_loss": 4.178745380946619, "tokens_seen": 280887296 }, { "epoch": 0.09, "learning_rate": 0.0004620561685055166, "loss": 3.1536, "theoretical_loss": 4.178629775915066, "tokens_seen": 280952832 }, { "epoch": 0.09, "learning_rate": 0.0004620461384152457, "loss": 3.0706, "theoretical_loss": 4.178514205395376, "tokens_seen": 281018368 }, { "epoch": 0.09, "learning_rate": 0.00046203610832497495, "loss": 3.0371, "theoretical_loss": 4.178398669369201, "tokens_seen": 281083904 }, { "epoch": 0.09, "learning_rate": 0.0004620260782347041, "loss": 2.8429, "theoretical_loss": 4.178283167818206, "tokens_seen": 281149440 }, { "epoch": 0.09, "learning_rate": 0.0004620160481444333, "loss": 3.1415, "theoretical_loss": 4.178167700724073, "tokens_seen": 281214976 }, { "epoch": 0.09, "learning_rate": 0.0004620060180541625, "loss": 3.0451, "theoretical_loss": 4.178052268068494, "tokens_seen": 281280512 }, { "epoch": 0.09, "learning_rate": 0.00046199598796389167, "loss": 3.0419, "theoretical_loss": 4.177936869833179, "tokens_seen": 281346048 }, { "epoch": 0.09, "learning_rate": 0.00046198595787362085, "loss": 2.8053, "theoretical_loss": 4.17782150599985, "tokens_seen": 281411584 }, { "epoch": 0.09, "learning_rate": 0.0004619759277833501, "loss": 2.7688, "theoretical_loss": 4.1777061765502435, "tokens_seen": 281477120 }, { "epoch": 0.09, "learning_rate": 0.0004619658976930792, "loss": 3.022, "theoretical_loss": 4.1775908814661085, "tokens_seen": 281542656 }, { "epoch": 0.09, "learning_rate": 0.00046195586760280845, "loss": 3.3437, "theoretical_loss": 4.17747562072921, "tokens_seen": 281608192 }, { "epoch": 0.09, "learning_rate": 0.00046194583751253763, "loss": 2.9559, "theoretical_loss": 4.177360394321325, "tokens_seen": 281673728 }, { "epoch": 0.09, "learning_rate": 0.0004619358074222668, "loss": 2.7982, "theoretical_loss": 4.177245202224246, "tokens_seen": 281739264 }, { "epoch": 0.09, "objective/train/docs_used": 171008, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.037534236907959, "objective/train/theoretical_loss": 4.17713004441978, "objective/train/tokens_used": 302264800, "theoretical_loss": 4.17713004441978, "tokens_seen": 281804800 }, { "epoch": 0.09, "learning_rate": 0.000461925777331996, "loss": 3.1612, "theoretical_loss": 4.17713004441978, "tokens_seen": 281804800 }, { "epoch": 0.09, "learning_rate": 0.0004619157472417252, "loss": 3.2436, "theoretical_loss": 4.177014920889745, "tokens_seen": 281870336 }, { "epoch": 0.09, "learning_rate": 0.00046190571715145435, "loss": 3.0578, "theoretical_loss": 4.176899831615974, "tokens_seen": 281935872 }, { "epoch": 0.09, "learning_rate": 0.0004618956870611836, "loss": 3.1397, "theoretical_loss": 4.176784776580316, "tokens_seen": 282001408 }, { "epoch": 0.09, "learning_rate": 0.0004618856569709127, "loss": 3.2312, "theoretical_loss": 4.176669755764632, "tokens_seen": 282066944 }, { "epoch": 0.09, "learning_rate": 0.00046187562688064195, "loss": 3.0992, "theoretical_loss": 4.176554769150796, "tokens_seen": 282132480 }, { "epoch": 0.09, "learning_rate": 0.0004618655967903711, "loss": 3.1607, "theoretical_loss": 4.176439816720697, "tokens_seen": 282198016 }, { "epoch": 0.09, "learning_rate": 0.0004618555667001003, "loss": 2.9453, "theoretical_loss": 4.1763248984562376, "tokens_seen": 282263552 }, { "epoch": 0.09, "learning_rate": 0.0004618455366098295, "loss": 3.2318, "theoretical_loss": 4.176210014339335, "tokens_seen": 282329088 }, { "epoch": 0.09, "learning_rate": 0.0004618355065195587, "loss": 3.171, "theoretical_loss": 4.17609516435192, "tokens_seen": 282394624 }, { "epoch": 0.09, "learning_rate": 0.00046182547642928786, "loss": 3.0733, "theoretical_loss": 4.1759803484759335, "tokens_seen": 282460160 }, { "epoch": 0.09, "learning_rate": 0.00046181544633901704, "loss": 3.0515, "theoretical_loss": 4.175865566693336, "tokens_seen": 282525696 }, { "epoch": 0.09, "learning_rate": 0.0004618054162487462, "loss": 3.2269, "theoretical_loss": 4.175750818986098, "tokens_seen": 282591232 }, { "epoch": 0.09, "learning_rate": 0.00046179538615847545, "loss": 3.2102, "theoretical_loss": 4.1756361053362046, "tokens_seen": 282656768 }, { "epoch": 0.09, "learning_rate": 0.0004617853560682046, "loss": 3.2833, "theoretical_loss": 4.1755214257256545, "tokens_seen": 282722304 }, { "epoch": 0.09, "learning_rate": 0.0004617753259779338, "loss": 2.9001, "theoretical_loss": 4.17540678013646, "tokens_seen": 282787840 }, { "epoch": 0.09, "learning_rate": 0.000461765295887663, "loss": 2.989, "theoretical_loss": 4.175292168550648, "tokens_seen": 282853376 }, { "epoch": 0.09, "learning_rate": 0.0004617552657973922, "loss": 3.2841, "theoretical_loss": 4.175177590950257, "tokens_seen": 282918912 }, { "epoch": 0.09, "learning_rate": 0.00046174523570712136, "loss": 2.9466, "theoretical_loss": 4.175063047317342, "tokens_seen": 282984448 }, { "epoch": 0.09, "learning_rate": 0.00046173520561685054, "loss": 2.8851, "theoretical_loss": 4.174948537633968, "tokens_seen": 283049984 }, { "epoch": 0.09, "learning_rate": 0.0004617251755265797, "loss": 3.0796, "theoretical_loss": 4.174834061882218, "tokens_seen": 283115520 }, { "epoch": 0.09, "learning_rate": 0.00046171514543630896, "loss": 3.0094, "theoretical_loss": 4.1747196200441845, "tokens_seen": 283181056 }, { "epoch": 0.09, "learning_rate": 0.00046170511534603814, "loss": 3.125, "theoretical_loss": 4.174605212101977, "tokens_seen": 283246592 }, { "epoch": 0.09, "learning_rate": 0.0004616950852557673, "loss": 3.1404, "theoretical_loss": 4.174490838037716, "tokens_seen": 283312128 }, { "epoch": 0.09, "learning_rate": 0.0004616850551654965, "loss": 3.1135, "theoretical_loss": 4.174376497833537, "tokens_seen": 283377664 }, { "epoch": 0.09, "objective/train/docs_used": 172282, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8163797855377197, "objective/train/theoretical_loss": 4.174262191471587, "objective/train/tokens_used": 303903200, "theoretical_loss": 4.174262191471587, "tokens_seen": 283443200 }, { "epoch": 0.09, "learning_rate": 0.0004616750250752257, "loss": 2.8791, "theoretical_loss": 4.174262191471587, "tokens_seen": 283443200 }, { "epoch": 0.09, "learning_rate": 0.0004616649949849549, "loss": 2.9439, "theoretical_loss": 4.17414791893403, "tokens_seen": 283508736 }, { "epoch": 0.09, "learning_rate": 0.00046165496489468404, "loss": 2.934, "theoretical_loss": 4.17403368020304, "tokens_seen": 283574272 }, { "epoch": 0.09, "learning_rate": 0.0004616449348044133, "loss": 3.2267, "theoretical_loss": 4.173919475260808, "tokens_seen": 283639808 }, { "epoch": 0.09, "learning_rate": 0.0004616349047141424, "loss": 3.2523, "theoretical_loss": 4.173805304089536, "tokens_seen": 283705344 }, { "epoch": 0.09, "learning_rate": 0.00046162487462387164, "loss": 3.0485, "theoretical_loss": 4.173691166671439, "tokens_seen": 283770880 }, { "epoch": 0.09, "learning_rate": 0.0004616148445336008, "loss": 3.3109, "theoretical_loss": 4.173577062988748, "tokens_seen": 283836416 }, { "epoch": 0.09, "learning_rate": 0.00046160481444333, "loss": 3.2738, "theoretical_loss": 4.173462993023706, "tokens_seen": 283901952 }, { "epoch": 0.09, "learning_rate": 0.0004615947843530592, "loss": 2.992, "theoretical_loss": 4.173348956758568, "tokens_seen": 283967488 }, { "epoch": 0.09, "learning_rate": 0.0004615847542627884, "loss": 2.9936, "theoretical_loss": 4.173234954175605, "tokens_seen": 284033024 }, { "epoch": 0.09, "learning_rate": 0.00046157472417251755, "loss": 3.3821, "theoretical_loss": 4.173120985257102, "tokens_seen": 284098560 }, { "epoch": 0.09, "learning_rate": 0.0004615646940822468, "loss": 2.9413, "theoretical_loss": 4.173007049985352, "tokens_seen": 284164096 }, { "epoch": 0.09, "learning_rate": 0.0004615546639919759, "loss": 2.9063, "theoretical_loss": 4.172893148342667, "tokens_seen": 284229632 }, { "epoch": 0.09, "learning_rate": 0.00046154463390170515, "loss": 3.2355, "theoretical_loss": 4.172779280311372, "tokens_seen": 284295168 }, { "epoch": 0.09, "learning_rate": 0.0004615346038114343, "loss": 3.1892, "theoretical_loss": 4.172665445873801, "tokens_seen": 284360704 }, { "epoch": 0.09, "learning_rate": 0.0004615245737211635, "loss": 3.0414, "theoretical_loss": 4.172551645012307, "tokens_seen": 284426240 }, { "epoch": 0.09, "learning_rate": 0.0004615145436308927, "loss": 2.9639, "theoretical_loss": 4.1724378777092515, "tokens_seen": 284491776 }, { "epoch": 0.09, "learning_rate": 0.00046150451354062187, "loss": 3.2605, "theoretical_loss": 4.172324143947012, "tokens_seen": 284557312 }, { "epoch": 0.09, "learning_rate": 0.00046149448345035105, "loss": 3.0332, "theoretical_loss": 4.172210443707979, "tokens_seen": 284622848 }, { "epoch": 0.09, "learning_rate": 0.0004614844533600803, "loss": 3.0535, "theoretical_loss": 4.1720967769745565, "tokens_seen": 284688384 }, { "epoch": 0.09, "learning_rate": 0.0004614744232698094, "loss": 2.8846, "theoretical_loss": 4.171983143729159, "tokens_seen": 284753920 }, { "epoch": 0.09, "learning_rate": 0.00046146439317953865, "loss": 3.3499, "theoretical_loss": 4.1718695439542195, "tokens_seen": 284819456 }, { "epoch": 0.09, "learning_rate": 0.00046145436308926783, "loss": 3.0726, "theoretical_loss": 4.17175597763218, "tokens_seen": 284884992 }, { "epoch": 0.09, "learning_rate": 0.000461444332998997, "loss": 2.7706, "theoretical_loss": 4.171642444745497, "tokens_seen": 284950528 }, { "epoch": 0.09, "learning_rate": 0.0004614343029087262, "loss": 3.0167, "theoretical_loss": 4.1715289452766395, "tokens_seen": 285016064 }, { "epoch": 0.09, "objective/train/docs_used": 172985, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.503798246383667, "objective/train/theoretical_loss": 4.1714154792080915, "objective/train/tokens_used": 305541600, "theoretical_loss": 4.1714154792080915, "tokens_seen": 285081600 }, { "epoch": 0.09, "learning_rate": 0.0004614242728184554, "loss": 3.1825, "theoretical_loss": 4.1714154792080915, "tokens_seen": 285081600 }, { "epoch": 0.09, "learning_rate": 0.00046141424272818455, "loss": 2.9299, "theoretical_loss": 4.171302046522349, "tokens_seen": 285147136 }, { "epoch": 0.09, "learning_rate": 0.0004614042126379138, "loss": 2.9893, "theoretical_loss": 4.171188647201921, "tokens_seen": 285212672 }, { "epoch": 0.09, "learning_rate": 0.0004613941825476429, "loss": 3.2218, "theoretical_loss": 4.1710752812293315, "tokens_seen": 285278208 }, { "epoch": 0.09, "learning_rate": 0.00046138415245737215, "loss": 3.0444, "theoretical_loss": 4.170961948587115, "tokens_seen": 285343744 }, { "epoch": 0.09, "learning_rate": 0.0004613741223671013, "loss": 3.0153, "theoretical_loss": 4.17084864925782, "tokens_seen": 285409280 }, { "epoch": 0.09, "learning_rate": 0.0004613640922768305, "loss": 3.0765, "theoretical_loss": 4.1707353832240095, "tokens_seen": 285474816 }, { "epoch": 0.09, "learning_rate": 0.0004613540621865597, "loss": 3.1904, "theoretical_loss": 4.170622150468258, "tokens_seen": 285540352 }, { "epoch": 0.09, "learning_rate": 0.0004613440320962889, "loss": 3.2306, "theoretical_loss": 4.170508950973154, "tokens_seen": 285605888 }, { "epoch": 0.09, "learning_rate": 0.00046133400200601806, "loss": 3.3067, "theoretical_loss": 4.1703957847213, "tokens_seen": 285671424 }, { "epoch": 0.09, "learning_rate": 0.00046132397191574724, "loss": 3.0881, "theoretical_loss": 4.170282651695308, "tokens_seen": 285736960 }, { "epoch": 0.09, "learning_rate": 0.0004613139418254764, "loss": 2.9596, "theoretical_loss": 4.170169551877808, "tokens_seen": 285802496 }, { "epoch": 0.09, "learning_rate": 0.00046130391173520566, "loss": 3.2447, "theoretical_loss": 4.170056485251439, "tokens_seen": 285868032 }, { "epoch": 0.09, "learning_rate": 0.0004612938816449348, "loss": 3.2723, "theoretical_loss": 4.169943451798856, "tokens_seen": 285933568 }, { "epoch": 0.09, "learning_rate": 0.000461283851554664, "loss": 3.0463, "theoretical_loss": 4.169830451502724, "tokens_seen": 285999104 }, { "epoch": 0.09, "learning_rate": 0.0004612738214643932, "loss": 2.924, "theoretical_loss": 4.169717484345725, "tokens_seen": 286064640 }, { "epoch": 0.09, "learning_rate": 0.0004612637913741224, "loss": 3.0269, "theoretical_loss": 4.1696045503105506, "tokens_seen": 286130176 }, { "epoch": 0.09, "learning_rate": 0.00046125376128385156, "loss": 3.1089, "theoretical_loss": 4.169491649379905, "tokens_seen": 286195712 }, { "epoch": 0.09, "learning_rate": 0.00046124373119358074, "loss": 3.0891, "theoretical_loss": 4.169378781536509, "tokens_seen": 286261248 }, { "epoch": 0.09, "learning_rate": 0.0004612337011033099, "loss": 3.0611, "theoretical_loss": 4.169265946763095, "tokens_seen": 286326784 }, { "epoch": 0.09, "learning_rate": 0.00046122367101303916, "loss": 3.0257, "theoretical_loss": 4.169153145042405, "tokens_seen": 286392320 }, { "epoch": 0.09, "learning_rate": 0.0004612136409227683, "loss": 3.1311, "theoretical_loss": 4.169040376357199, "tokens_seen": 286457856 }, { "epoch": 0.09, "learning_rate": 0.0004612036108324975, "loss": 3.2128, "theoretical_loss": 4.168927640690246, "tokens_seen": 286523392 }, { "epoch": 0.09, "learning_rate": 0.00046119358074222665, "loss": 2.8733, "theoretical_loss": 4.16881493802433, "tokens_seen": 286588928 }, { "epoch": 0.09, "learning_rate": 0.0004611835506519559, "loss": 2.9867, "theoretical_loss": 4.168702268342248, "tokens_seen": 286654464 }, { "epoch": 0.09, "objective/train/docs_used": 174327, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1925644874572754, "objective/train/theoretical_loss": 4.168589631626808, "objective/train/tokens_used": 307180000, "theoretical_loss": 4.168589631626808, "tokens_seen": 286720000 }, { "epoch": 0.09, "learning_rate": 0.00046117352056168506, "loss": 3.043, "theoretical_loss": 4.168589631626808, "tokens_seen": 286720000 }, { "epoch": 0.09, "learning_rate": 0.00046116349047141425, "loss": 2.8574, "theoretical_loss": 4.168477027860833, "tokens_seen": 286785536 }, { "epoch": 0.09, "learning_rate": 0.0004611534603811434, "loss": 3.0897, "theoretical_loss": 4.168364457027158, "tokens_seen": 286851072 }, { "epoch": 0.09, "learning_rate": 0.0004611434302908726, "loss": 3.0685, "theoretical_loss": 4.168251919108632, "tokens_seen": 286916608 }, { "epoch": 0.09, "learning_rate": 0.0004611334002006018, "loss": 3.356, "theoretical_loss": 4.168139414088113, "tokens_seen": 286982144 }, { "epoch": 0.09, "learning_rate": 0.000461123370110331, "loss": 3.1332, "theoretical_loss": 4.168026941948478, "tokens_seen": 287047680 }, { "epoch": 0.09, "learning_rate": 0.00046111334002006015, "loss": 3.1427, "theoretical_loss": 4.167914502672611, "tokens_seen": 287113216 }, { "epoch": 0.09, "learning_rate": 0.0004611033099297894, "loss": 3.263, "theoretical_loss": 4.1678020962434115, "tokens_seen": 287178752 }, { "epoch": 0.09, "learning_rate": 0.00046109327983951857, "loss": 2.847, "theoretical_loss": 4.167689722643792, "tokens_seen": 287244288 }, { "epoch": 0.09, "learning_rate": 0.00046108324974924775, "loss": 3.1922, "theoretical_loss": 4.1675773818566775, "tokens_seen": 287309824 }, { "epoch": 0.09, "learning_rate": 0.00046107321965897693, "loss": 3.224, "theoretical_loss": 4.167465073865006, "tokens_seen": 287375360 }, { "epoch": 0.09, "learning_rate": 0.0004610631895687061, "loss": 3.2108, "theoretical_loss": 4.167352798651726, "tokens_seen": 287440896 }, { "epoch": 0.09, "learning_rate": 0.0004610531594784353, "loss": 2.9368, "theoretical_loss": 4.167240556199802, "tokens_seen": 287506432 }, { "epoch": 0.09, "learning_rate": 0.0004610431293881645, "loss": 3.025, "theoretical_loss": 4.167128346492211, "tokens_seen": 287571968 }, { "epoch": 0.09, "learning_rate": 0.00046103309929789365, "loss": 3.1749, "theoretical_loss": 4.16701616951194, "tokens_seen": 287637504 }, { "epoch": 0.09, "learning_rate": 0.0004610230692076229, "loss": 3.2276, "theoretical_loss": 4.1669040252419896, "tokens_seen": 287703040 }, { "epoch": 0.09, "learning_rate": 0.000461013039117352, "loss": 3.1166, "theoretical_loss": 4.166791913665375, "tokens_seen": 287768576 }, { "epoch": 0.09, "learning_rate": 0.00046100300902708125, "loss": 3.2476, "theoretical_loss": 4.166679834765123, "tokens_seen": 287834112 }, { "epoch": 0.09, "learning_rate": 0.00046099297893681043, "loss": 3.2406, "theoretical_loss": 4.166567788524272, "tokens_seen": 287899648 }, { "epoch": 0.09, "learning_rate": 0.0004609829488465396, "loss": 3.1236, "theoretical_loss": 4.166455774925875, "tokens_seen": 287965184 }, { "epoch": 0.09, "learning_rate": 0.0004609729187562688, "loss": 2.6893, "theoretical_loss": 4.166343793952995, "tokens_seen": 288030720 }, { "epoch": 0.09, "learning_rate": 0.00046096288866599803, "loss": 2.9429, "theoretical_loss": 4.166231845588712, "tokens_seen": 288096256 }, { "epoch": 0.09, "learning_rate": 0.0004609528585757272, "loss": 3.1807, "theoretical_loss": 4.166119929816113, "tokens_seen": 288161792 }, { "epoch": 0.09, "learning_rate": 0.0004609428284854564, "loss": 3.054, "theoretical_loss": 4.166008046618303, "tokens_seen": 288227328 }, { "epoch": 0.09, "learning_rate": 0.0004609327983951856, "loss": 3.1462, "theoretical_loss": 4.1658961959783944, "tokens_seen": 288292864 }, { "epoch": 0.09, "objective/train/docs_used": 174865, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1643199920654297, "objective/train/theoretical_loss": 4.165784377879517, "objective/train/tokens_used": 308818400, "theoretical_loss": 4.165784377879517, "tokens_seen": 288358400 }, { "epoch": 0.09, "learning_rate": 0.00046092276830491475, "loss": 3.2366, "theoretical_loss": 4.165784377879517, "tokens_seen": 288358400 }, { "epoch": 0.09, "learning_rate": 0.000460912738214644, "loss": 2.9021, "theoretical_loss": 4.165672592304811, "tokens_seen": 288423936 }, { "epoch": 0.09, "learning_rate": 0.0004609027081243731, "loss": 3.0573, "theoretical_loss": 4.165560839237429, "tokens_seen": 288489472 }, { "epoch": 0.09, "learning_rate": 0.00046089267803410235, "loss": 3.0862, "theoretical_loss": 4.165449118660536, "tokens_seen": 288555008 }, { "epoch": 0.09, "learning_rate": 0.0004608826479438315, "loss": 3.149, "theoretical_loss": 4.16533743055731, "tokens_seen": 288620544 }, { "epoch": 0.09, "learning_rate": 0.0004608726178535607, "loss": 3.21, "theoretical_loss": 4.165225774910941, "tokens_seen": 288686080 }, { "epoch": 0.09, "learning_rate": 0.0004608625877632899, "loss": 2.8823, "theoretical_loss": 4.165114151704634, "tokens_seen": 288751616 }, { "epoch": 0.09, "learning_rate": 0.0004608525576730191, "loss": 3.0306, "theoretical_loss": 4.165002560921601, "tokens_seen": 288817152 }, { "epoch": 0.09, "learning_rate": 0.00046084252758274826, "loss": 3.2771, "theoretical_loss": 4.164891002545073, "tokens_seen": 288882688 }, { "epoch": 0.09, "learning_rate": 0.00046083249749247744, "loss": 2.7558, "theoretical_loss": 4.16477947655829, "tokens_seen": 288948224 }, { "epoch": 0.09, "learning_rate": 0.0004608224674022066, "loss": 3.0144, "theoretical_loss": 4.164667982944504, "tokens_seen": 289013760 }, { "epoch": 0.09, "learning_rate": 0.00046081243731193586, "loss": 2.9461, "theoretical_loss": 4.164556521686981, "tokens_seen": 289079296 }, { "epoch": 0.09, "learning_rate": 0.000460802407221665, "loss": 3.1734, "theoretical_loss": 4.1644450927689975, "tokens_seen": 289144832 }, { "epoch": 0.09, "learning_rate": 0.0004607923771313942, "loss": 3.2232, "theoretical_loss": 4.164333696173846, "tokens_seen": 289210368 }, { "epoch": 0.09, "learning_rate": 0.0004607823470411234, "loss": 3.0528, "theoretical_loss": 4.164222331884827, "tokens_seen": 289275904 }, { "epoch": 0.09, "learning_rate": 0.0004607723169508526, "loss": 3.0869, "theoretical_loss": 4.164110999885256, "tokens_seen": 289341440 }, { "epoch": 0.09, "learning_rate": 0.00046076228686058176, "loss": 3.0237, "theoretical_loss": 4.163999700158462, "tokens_seen": 289406976 }, { "epoch": 0.09, "learning_rate": 0.00046075225677031094, "loss": 3.0699, "theoretical_loss": 4.163888432687784, "tokens_seen": 289472512 }, { "epoch": 0.09, "learning_rate": 0.0004607422266800401, "loss": 2.7908, "theoretical_loss": 4.163777197456573, "tokens_seen": 289538048 }, { "epoch": 0.09, "learning_rate": 0.00046073219658976936, "loss": 3.0935, "theoretical_loss": 4.163665994448197, "tokens_seen": 289603584 }, { "epoch": 0.09, "learning_rate": 0.0004607221664994985, "loss": 2.9748, "theoretical_loss": 4.163554823646027, "tokens_seen": 289669120 }, { "epoch": 0.09, "learning_rate": 0.0004607121364092277, "loss": 3.0475, "theoretical_loss": 4.163443685033458, "tokens_seen": 289734656 }, { "epoch": 0.09, "learning_rate": 0.00046070210631895685, "loss": 2.9678, "theoretical_loss": 4.163332578593889, "tokens_seen": 289800192 }, { "epoch": 0.09, "learning_rate": 0.0004606920762286861, "loss": 3.0884, "theoretical_loss": 4.163221504310734, "tokens_seen": 289865728 }, { "epoch": 0.09, "learning_rate": 0.00046068204613841526, "loss": 3.1743, "theoretical_loss": 4.1631104621674195, "tokens_seen": 289931264 }, { "epoch": 0.09, "objective/train/docs_used": 176137, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.816805124282837, "objective/train/theoretical_loss": 4.162999452147384, "objective/train/tokens_used": 310456800, "theoretical_loss": 4.162999452147384, "tokens_seen": 289996800 }, { "epoch": 0.09, "learning_rate": 0.00046067201604814445, "loss": 2.9639, "theoretical_loss": 4.162999452147384, "tokens_seen": 289996800 }, { "epoch": 0.09, "learning_rate": 0.0004606619859578736, "loss": 2.9778, "theoretical_loss": 4.1628884742340775, "tokens_seen": 290062336 }, { "epoch": 0.09, "learning_rate": 0.0004606519558676028, "loss": 3.0018, "theoretical_loss": 4.162777528410963, "tokens_seen": 290127872 }, { "epoch": 0.09, "learning_rate": 0.000460641925777332, "loss": 2.9703, "theoretical_loss": 4.162666614661518, "tokens_seen": 290193408 }, { "epoch": 0.09, "learning_rate": 0.0004606318956870612, "loss": 3.2376, "theoretical_loss": 4.162555732969227, "tokens_seen": 290258944 }, { "epoch": 0.09, "learning_rate": 0.00046062186559679035, "loss": 3.0843, "theoretical_loss": 4.162444883317591, "tokens_seen": 290324480 }, { "epoch": 0.09, "learning_rate": 0.0004606118355065196, "loss": 3.1161, "theoretical_loss": 4.162334065690123, "tokens_seen": 290390016 }, { "epoch": 0.09, "learning_rate": 0.00046060180541624877, "loss": 3.0975, "theoretical_loss": 4.162223280070345, "tokens_seen": 290455552 }, { "epoch": 0.09, "learning_rate": 0.00046059177532597795, "loss": 3.0539, "theoretical_loss": 4.1621125264417955, "tokens_seen": 290521088 }, { "epoch": 0.09, "learning_rate": 0.00046058174523570713, "loss": 3.2598, "theoretical_loss": 4.162001804788021, "tokens_seen": 290586624 }, { "epoch": 0.09, "learning_rate": 0.0004605717151454363, "loss": 3.1444, "theoretical_loss": 4.161891115092583, "tokens_seen": 290652160 }, { "epoch": 0.09, "learning_rate": 0.0004605616850551655, "loss": 3.164, "theoretical_loss": 4.161780457339055, "tokens_seen": 290717696 }, { "epoch": 0.09, "learning_rate": 0.0004605516549648947, "loss": 3.1421, "theoretical_loss": 4.161669831511022, "tokens_seen": 290783232 }, { "epoch": 0.09, "learning_rate": 0.00046054162487462385, "loss": 3.167, "theoretical_loss": 4.16155923759208, "tokens_seen": 290848768 }, { "epoch": 0.09, "learning_rate": 0.0004605315947843531, "loss": 3.0247, "theoretical_loss": 4.161448675565838, "tokens_seen": 290914304 }, { "epoch": 0.09, "learning_rate": 0.0004605215646940822, "loss": 3.2994, "theoretical_loss": 4.161338145415918, "tokens_seen": 290979840 }, { "epoch": 0.09, "learning_rate": 0.00046051153460381145, "loss": 2.9411, "theoretical_loss": 4.161227647125955, "tokens_seen": 291045376 }, { "epoch": 0.09, "learning_rate": 0.00046050150451354063, "loss": 2.9685, "theoretical_loss": 4.161117180679591, "tokens_seen": 291110912 }, { "epoch": 0.09, "learning_rate": 0.0004604914744232698, "loss": 2.8291, "theoretical_loss": 4.161006746060488, "tokens_seen": 291176448 }, { "epoch": 0.09, "learning_rate": 0.000460481444332999, "loss": 3.0402, "theoretical_loss": 4.160896343252311, "tokens_seen": 291241984 }, { "epoch": 0.09, "learning_rate": 0.00046047141424272823, "loss": 3.0417, "theoretical_loss": 4.160785972238745, "tokens_seen": 291307520 }, { "epoch": 0.09, "learning_rate": 0.00046046138415245736, "loss": 3.3467, "theoretical_loss": 4.160675633003484, "tokens_seen": 291373056 }, { "epoch": 0.09, "learning_rate": 0.0004604513540621866, "loss": 3.4501, "theoretical_loss": 4.16056532553023, "tokens_seen": 291438592 }, { "epoch": 0.09, "learning_rate": 0.0004604413239719157, "loss": 2.9161, "theoretical_loss": 4.160455049802706, "tokens_seen": 291504128 }, { "epoch": 0.09, "learning_rate": 0.00046043129388164495, "loss": 3.0043, "theoretical_loss": 4.1603448058046375, "tokens_seen": 291569664 }, { "epoch": 0.09, "objective/train/docs_used": 176478, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.099493980407715, "objective/train/theoretical_loss": 4.160234593519768, "objective/train/tokens_used": 312095200, "theoretical_loss": 4.160234593519768, "tokens_seen": 291635200 }, { "epoch": 0.09, "learning_rate": 0.00046042126379137414, "loss": 3.4061, "theoretical_loss": 4.160234593519768, "tokens_seen": 291635200 }, { "epoch": 0.09, "learning_rate": 0.0004604112337011033, "loss": 2.9333, "theoretical_loss": 4.160124412931852, "tokens_seen": 291700736 }, { "epoch": 0.09, "learning_rate": 0.0004604012036108325, "loss": 3.2184, "theoretical_loss": 4.160014264024654, "tokens_seen": 291766272 }, { "epoch": 0.09, "learning_rate": 0.0004603911735205617, "loss": 3.0181, "theoretical_loss": 4.159904146781952, "tokens_seen": 291831808 }, { "epoch": 0.09, "learning_rate": 0.00046038114343029086, "loss": 2.9699, "theoretical_loss": 4.159794061187536, "tokens_seen": 291897344 }, { "epoch": 0.09, "learning_rate": 0.0004603711133400201, "loss": 3.0901, "theoretical_loss": 4.1596840072252075, "tokens_seen": 291962880 }, { "epoch": 0.09, "learning_rate": 0.0004603610832497492, "loss": 3.0625, "theoretical_loss": 4.159573984878779, "tokens_seen": 292028416 }, { "epoch": 0.09, "learning_rate": 0.00046035105315947846, "loss": 3.0933, "theoretical_loss": 4.159463994132079, "tokens_seen": 292093952 }, { "epoch": 0.09, "learning_rate": 0.0004603410230692076, "loss": 3.1878, "theoretical_loss": 4.15935403496894, "tokens_seen": 292159488 }, { "epoch": 0.09, "learning_rate": 0.0004603309929789368, "loss": 3.2494, "theoretical_loss": 4.159244107373215, "tokens_seen": 292225024 }, { "epoch": 0.09, "learning_rate": 0.000460320962888666, "loss": 3.1238, "theoretical_loss": 4.159134211328765, "tokens_seen": 292290560 }, { "epoch": 0.09, "learning_rate": 0.0004603109327983952, "loss": 3.2343, "theoretical_loss": 4.159024346819461, "tokens_seen": 292356096 }, { "epoch": 0.09, "learning_rate": 0.00046030090270812436, "loss": 2.9651, "theoretical_loss": 4.158914513829189, "tokens_seen": 292421632 }, { "epoch": 0.09, "learning_rate": 0.0004602908726178536, "loss": 3.1782, "theoretical_loss": 4.158804712341845, "tokens_seen": 292487168 }, { "epoch": 0.09, "learning_rate": 0.0004602808425275827, "loss": 3.2679, "theoretical_loss": 4.158694942341338, "tokens_seen": 292552704 }, { "epoch": 0.09, "learning_rate": 0.00046027081243731196, "loss": 3.0873, "theoretical_loss": 4.1585852038115885, "tokens_seen": 292618240 }, { "epoch": 0.09, "learning_rate": 0.0004602607823470411, "loss": 3.0467, "theoretical_loss": 4.1584754967365285, "tokens_seen": 292683776 }, { "epoch": 0.09, "learning_rate": 0.0004602507522567703, "loss": 3.2751, "theoretical_loss": 4.1583658211001016, "tokens_seen": 292749312 }, { "epoch": 0.09, "learning_rate": 0.0004602407221664995, "loss": 3.3058, "theoretical_loss": 4.158256176886264, "tokens_seen": 292814848 }, { "epoch": 0.09, "learning_rate": 0.0004602306920762287, "loss": 3.0265, "theoretical_loss": 4.158146564078982, "tokens_seen": 292880384 }, { "epoch": 0.09, "learning_rate": 0.00046022066198595787, "loss": 3.0768, "theoretical_loss": 4.158036982662237, "tokens_seen": 292945920 }, { "epoch": 0.09, "learning_rate": 0.00046021063189568705, "loss": 2.8461, "theoretical_loss": 4.157927432620018, "tokens_seen": 293011456 }, { "epoch": 0.09, "learning_rate": 0.0004602006018054163, "loss": 3.1769, "theoretical_loss": 4.157817913936329, "tokens_seen": 293076992 }, { "epoch": 0.09, "learning_rate": 0.00046019057171514546, "loss": 2.9548, "theoretical_loss": 4.157708426595184, "tokens_seen": 293142528 }, { "epoch": 0.09, "learning_rate": 0.00046018054162487465, "loss": 2.9129, "theoretical_loss": 4.157598970580608, "tokens_seen": 293208064 }, { "epoch": 0.09, "objective/train/docs_used": 177825, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.218212842941284, "objective/train/theoretical_loss": 4.157489545876642, "objective/train/tokens_used": 313733600, "theoretical_loss": 4.157489545876642, "tokens_seen": 293273600 }, { "epoch": 0.09, "learning_rate": 0.0004601705115346038, "loss": 3.1663, "theoretical_loss": 4.157489545876642, "tokens_seen": 293273600 }, { "epoch": 0.09, "learning_rate": 0.000460160481444333, "loss": 3.228, "theoretical_loss": 4.157380152467333, "tokens_seen": 293339136 }, { "epoch": 0.09, "learning_rate": 0.0004601504513540622, "loss": 3.1855, "theoretical_loss": 4.157270790336742, "tokens_seen": 293404672 }, { "epoch": 0.09, "learning_rate": 0.0004601404212637914, "loss": 3.1031, "theoretical_loss": 4.157161459468944, "tokens_seen": 293470208 }, { "epoch": 0.09, "learning_rate": 0.00046013039117352055, "loss": 3.11, "theoretical_loss": 4.157052159848023, "tokens_seen": 293535744 }, { "epoch": 0.09, "learning_rate": 0.0004601203610832498, "loss": 2.8797, "theoretical_loss": 4.156942891458074, "tokens_seen": 293601280 }, { "epoch": 0.09, "learning_rate": 0.00046011033099297897, "loss": 3.1523, "theoretical_loss": 4.156833654283207, "tokens_seen": 293666816 }, { "epoch": 0.09, "learning_rate": 0.00046010030090270815, "loss": 2.9046, "theoretical_loss": 4.15672444830754, "tokens_seen": 293732352 }, { "epoch": 0.09, "learning_rate": 0.00046009027081243733, "loss": 3.1712, "theoretical_loss": 4.156615273515205, "tokens_seen": 293797888 }, { "epoch": 0.09, "learning_rate": 0.0004600802407221665, "loss": 3.1373, "theoretical_loss": 4.156506129890344, "tokens_seen": 293863424 }, { "epoch": 0.09, "learning_rate": 0.0004600702106318957, "loss": 3.2344, "theoretical_loss": 4.156397017417111, "tokens_seen": 293928960 }, { "epoch": 0.09, "learning_rate": 0.00046006018054162493, "loss": 3.2112, "theoretical_loss": 4.156287936079675, "tokens_seen": 293994496 }, { "epoch": 0.09, "learning_rate": 0.00046005015045135405, "loss": 3.186, "theoretical_loss": 4.156178885862209, "tokens_seen": 294060032 }, { "epoch": 0.09, "learning_rate": 0.0004600401203610833, "loss": 2.9581, "theoretical_loss": 4.156069866748906, "tokens_seen": 294125568 }, { "epoch": 0.09, "learning_rate": 0.0004600300902708124, "loss": 3.0633, "theoretical_loss": 4.155960878723965, "tokens_seen": 294191104 }, { "epoch": 0.09, "learning_rate": 0.00046002006018054165, "loss": 2.9769, "theoretical_loss": 4.155851921771598, "tokens_seen": 294256640 }, { "epoch": 0.09, "learning_rate": 0.00046001003009027083, "loss": 3.154, "theoretical_loss": 4.155742995876029, "tokens_seen": 294322176 }, { "epoch": 0.09, "learning_rate": 0.00046, "loss": 3.2509, "theoretical_loss": 4.155634101021494, "tokens_seen": 294387712 }, { "epoch": 0.09, "learning_rate": 0.0004599899699097292, "loss": 2.9986, "theoretical_loss": 4.155525237192238, "tokens_seen": 294453248 }, { "epoch": 0.09, "learning_rate": 0.00045997993981945843, "loss": 3.0416, "theoretical_loss": 4.155416404372522, "tokens_seen": 294518784 }, { "epoch": 0.09, "learning_rate": 0.00045996990972918756, "loss": 3.113, "theoretical_loss": 4.155307602546614, "tokens_seen": 294584320 }, { "epoch": 0.09, "learning_rate": 0.0004599598796389168, "loss": 3.1698, "theoretical_loss": 4.155198831698795, "tokens_seen": 294649856 }, { "epoch": 0.09, "learning_rate": 0.0004599498495486459, "loss": 3.1415, "theoretical_loss": 4.155090091813358, "tokens_seen": 294715392 }, { "epoch": 0.09, "learning_rate": 0.00045993981945837515, "loss": 2.9568, "theoretical_loss": 4.154981382874608, "tokens_seen": 294780928 }, { "epoch": 0.09, "learning_rate": 0.00045992978936810434, "loss": 3.0327, "theoretical_loss": 4.154872704866859, "tokens_seen": 294846464 }, { "epoch": 0.09, "objective/train/docs_used": 179027, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5512094497680664, "objective/train/theoretical_loss": 4.15476405777444, "objective/train/tokens_used": 315372000, "theoretical_loss": 4.15476405777444, "tokens_seen": 294912000 }, { "epoch": 0.09, "learning_rate": 0.0004599197592778335, "loss": 2.9603, "theoretical_loss": 4.15476405777444, "tokens_seen": 294912000 }, { "epoch": 0.09, "learning_rate": 0.0004599097291875627, "loss": 3.1664, "theoretical_loss": 4.154655441581687, "tokens_seen": 294977536 }, { "epoch": 0.09, "learning_rate": 0.0004598996990972919, "loss": 2.91, "theoretical_loss": 4.154546856272952, "tokens_seen": 295043072 }, { "epoch": 0.09, "learning_rate": 0.00045988966900702106, "loss": 2.9691, "theoretical_loss": 4.154438301832596, "tokens_seen": 295108608 }, { "epoch": 0.09, "learning_rate": 0.0004598796389167503, "loss": 3.0933, "theoretical_loss": 4.154329778244991, "tokens_seen": 295174144 }, { "epoch": 0.09, "learning_rate": 0.0004598696088264794, "loss": 2.9358, "theoretical_loss": 4.154221285494521, "tokens_seen": 295239680 }, { "epoch": 0.09, "learning_rate": 0.00045985957873620866, "loss": 3.2587, "theoretical_loss": 4.154112823565582, "tokens_seen": 295305216 }, { "epoch": 0.09, "learning_rate": 0.0004598495486459378, "loss": 3.3291, "theoretical_loss": 4.15400439244258, "tokens_seen": 295370752 }, { "epoch": 0.09, "learning_rate": 0.000459839518555667, "loss": 3.3911, "theoretical_loss": 4.153895992109935, "tokens_seen": 295436288 }, { "epoch": 0.09, "learning_rate": 0.0004598294884653962, "loss": 2.9428, "theoretical_loss": 4.153787622552073, "tokens_seen": 295501824 }, { "epoch": 0.09, "learning_rate": 0.0004598194583751254, "loss": 3.1373, "theoretical_loss": 4.153679283753439, "tokens_seen": 295567360 }, { "epoch": 0.09, "learning_rate": 0.00045980942828485456, "loss": 3.1252, "theoretical_loss": 4.15357097569848, "tokens_seen": 295632896 }, { "epoch": 0.09, "learning_rate": 0.0004597993981945838, "loss": 3.1504, "theoretical_loss": 4.153462698371665, "tokens_seen": 295698432 }, { "epoch": 0.09, "learning_rate": 0.0004597893681043129, "loss": 3.1088, "theoretical_loss": 4.1533544517574645, "tokens_seen": 295763968 }, { "epoch": 0.09, "learning_rate": 0.00045977933801404216, "loss": 2.9716, "theoretical_loss": 4.153246235840367, "tokens_seen": 295829504 }, { "epoch": 0.09, "learning_rate": 0.0004597693079237713, "loss": 3.2071, "theoretical_loss": 4.153138050604868, "tokens_seen": 295895040 }, { "epoch": 0.09, "learning_rate": 0.0004597592778335005, "loss": 3.2361, "theoretical_loss": 4.153029896035476, "tokens_seen": 295960576 }, { "epoch": 0.09, "learning_rate": 0.0004597492477432297, "loss": 2.9966, "theoretical_loss": 4.152921772116712, "tokens_seen": 296026112 }, { "epoch": 0.09, "learning_rate": 0.0004597392176529589, "loss": 3.1588, "theoretical_loss": 4.152813678833106, "tokens_seen": 296091648 }, { "epoch": 0.09, "learning_rate": 0.00045972918756268807, "loss": 3.2001, "theoretical_loss": 4.152705616169202, "tokens_seen": 296157184 }, { "epoch": 0.09, "learning_rate": 0.00045971915747241725, "loss": 2.8042, "theoretical_loss": 4.15259758410955, "tokens_seen": 296222720 }, { "epoch": 0.09, "learning_rate": 0.00045970912738214643, "loss": 3.135, "theoretical_loss": 4.152489582638719, "tokens_seen": 296288256 }, { "epoch": 0.09, "learning_rate": 0.00045969909729187566, "loss": 3.3503, "theoretical_loss": 4.152381611741281, "tokens_seen": 296353792 }, { "epoch": 0.09, "learning_rate": 0.0004596890672016048, "loss": 3.0074, "theoretical_loss": 4.152273671401824, "tokens_seen": 296419328 }, { "epoch": 0.09, "learning_rate": 0.000459679037111334, "loss": 3.1205, "theoretical_loss": 4.152165761604948, "tokens_seen": 296484864 }, { "epoch": 0.09, "objective/train/docs_used": 179648, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.256779432296753, "objective/train/theoretical_loss": 4.152057882335261, "objective/train/tokens_used": 317010400, "theoretical_loss": 4.152057882335261, "tokens_seen": 296550400 }, { "epoch": 0.09, "learning_rate": 0.00045966900702106315, "loss": 3.1838, "theoretical_loss": 4.152057882335261, "tokens_seen": 296550400 }, { "epoch": 0.09, "learning_rate": 0.0004596589769307924, "loss": 3.178, "theoretical_loss": 4.151950033577383, "tokens_seen": 296615936 }, { "epoch": 0.09, "learning_rate": 0.00045964894684052157, "loss": 3.1013, "theoretical_loss": 4.151842215315947, "tokens_seen": 296681472 }, { "epoch": 0.09, "learning_rate": 0.00045963891675025075, "loss": 3.056, "theoretical_loss": 4.151734427535594, "tokens_seen": 296747008 }, { "epoch": 0.09, "learning_rate": 0.00045962888665997993, "loss": 3.1372, "theoretical_loss": 4.151626670220979, "tokens_seen": 296812544 }, { "epoch": 0.09, "learning_rate": 0.00045961885656970917, "loss": 3.0696, "theoretical_loss": 4.151518943356768, "tokens_seen": 296878080 }, { "epoch": 0.09, "learning_rate": 0.0004596088264794383, "loss": 3.0229, "theoretical_loss": 4.151411246927636, "tokens_seen": 296943616 }, { "epoch": 0.09, "learning_rate": 0.00045959879638916753, "loss": 2.9934, "theoretical_loss": 4.15130358091827, "tokens_seen": 297009152 }, { "epoch": 0.09, "learning_rate": 0.00045958876629889666, "loss": 3.1986, "theoretical_loss": 4.151195945313369, "tokens_seen": 297074688 }, { "epoch": 0.09, "learning_rate": 0.0004595787362086259, "loss": 3.0798, "theoretical_loss": 4.151088340097642, "tokens_seen": 297140224 }, { "epoch": 0.09, "learning_rate": 0.0004595687061183551, "loss": 3.0475, "theoretical_loss": 4.15098076525581, "tokens_seen": 297205760 }, { "epoch": 0.09, "learning_rate": 0.00045955867602808425, "loss": 3.1249, "theoretical_loss": 4.150873220772604, "tokens_seen": 297271296 }, { "epoch": 0.09, "learning_rate": 0.00045954864593781344, "loss": 3.3091, "theoretical_loss": 4.150765706632766, "tokens_seen": 297336832 }, { "epoch": 0.09, "learning_rate": 0.0004595386158475426, "loss": 3.2366, "theoretical_loss": 4.1506582228210505, "tokens_seen": 297402368 }, { "epoch": 0.09, "learning_rate": 0.0004595285857572718, "loss": 2.9709, "theoretical_loss": 4.150550769322221, "tokens_seen": 297467904 }, { "epoch": 0.09, "learning_rate": 0.00045951855566700103, "loss": 3.1059, "theoretical_loss": 4.150443346121054, "tokens_seen": 297533440 }, { "epoch": 0.09, "learning_rate": 0.00045950852557673016, "loss": 2.9407, "theoretical_loss": 4.150335953202336, "tokens_seen": 297598976 }, { "epoch": 0.09, "learning_rate": 0.0004594984954864594, "loss": 3.0818, "theoretical_loss": 4.150228590550864, "tokens_seen": 297664512 }, { "epoch": 0.09, "learning_rate": 0.0004594884653961885, "loss": 3.2076, "theoretical_loss": 4.150121258151447, "tokens_seen": 297730048 }, { "epoch": 0.09, "learning_rate": 0.00045947843530591776, "loss": 2.9442, "theoretical_loss": 4.150013955988905, "tokens_seen": 297795584 }, { "epoch": 0.09, "learning_rate": 0.00045946840521564694, "loss": 3.1901, "theoretical_loss": 4.149906684048068, "tokens_seen": 297861120 }, { "epoch": 0.09, "learning_rate": 0.0004594583751253761, "loss": 2.6943, "theoretical_loss": 4.1497994423137765, "tokens_seen": 297926656 }, { "epoch": 0.09, "learning_rate": 0.00045944834503510536, "loss": 2.9976, "theoretical_loss": 4.149692230770884, "tokens_seen": 297992192 }, { "epoch": 0.09, "learning_rate": 0.00045943831494483454, "loss": 3.1648, "theoretical_loss": 4.149585049404253, "tokens_seen": 298057728 }, { "epoch": 0.09, "learning_rate": 0.0004594282848545637, "loss": 3.1697, "theoretical_loss": 4.149477898198759, "tokens_seen": 298123264 }, { "epoch": 0.09, "objective/train/docs_used": 180307, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1973071098327637, "objective/train/theoretical_loss": 4.149370777139286, "objective/train/tokens_used": 318648800, "theoretical_loss": 4.149370777139286, "tokens_seen": 298188800 }, { "epoch": 0.09, "learning_rate": 0.0004594182547642929, "loss": 3.0318, "theoretical_loss": 4.149370777139286, "tokens_seen": 298188800 }, { "epoch": 0.09, "learning_rate": 0.0004594082246740221, "loss": 3.1242, "theoretical_loss": 4.14926368621073, "tokens_seen": 298254336 }, { "epoch": 0.09, "learning_rate": 0.00045939819458375126, "loss": 3.0613, "theoretical_loss": 4.149156625397998, "tokens_seen": 298319872 }, { "epoch": 0.09, "learning_rate": 0.0004593881644934805, "loss": 2.8588, "theoretical_loss": 4.149049594686008, "tokens_seen": 298385408 }, { "epoch": 0.09, "learning_rate": 0.0004593781344032096, "loss": 2.7734, "theoretical_loss": 4.1489425940596885, "tokens_seen": 298450944 }, { "epoch": 0.09, "learning_rate": 0.00045936810431293886, "loss": 2.7681, "theoretical_loss": 4.148835623503978, "tokens_seen": 298516480 }, { "epoch": 0.09, "learning_rate": 0.000459358074222668, "loss": 3.1072, "theoretical_loss": 4.148728683003829, "tokens_seen": 298582016 }, { "epoch": 0.09, "learning_rate": 0.0004593480441323972, "loss": 3.1108, "theoretical_loss": 4.1486217725442005, "tokens_seen": 298647552 }, { "epoch": 0.09, "learning_rate": 0.0004593380140421264, "loss": 2.91, "theoretical_loss": 4.148514892110065, "tokens_seen": 298713088 }, { "epoch": 0.09, "learning_rate": 0.0004593279839518556, "loss": 3.2009, "theoretical_loss": 4.148408041686406, "tokens_seen": 298778624 }, { "epoch": 0.09, "learning_rate": 0.00045931795386158476, "loss": 3.0153, "theoretical_loss": 4.148301221258217, "tokens_seen": 298844160 }, { "epoch": 0.09, "learning_rate": 0.000459307923771314, "loss": 3.2481, "theoretical_loss": 4.148194430810502, "tokens_seen": 298909696 }, { "epoch": 0.09, "learning_rate": 0.0004592978936810431, "loss": 3.3113, "theoretical_loss": 4.148087670328276, "tokens_seen": 298975232 }, { "epoch": 0.09, "learning_rate": 0.00045928786359077236, "loss": 3.2193, "theoretical_loss": 4.147980939796565, "tokens_seen": 299040768 }, { "epoch": 0.09, "learning_rate": 0.0004592778335005015, "loss": 3.0393, "theoretical_loss": 4.147874239200405, "tokens_seen": 299106304 }, { "epoch": 0.09, "learning_rate": 0.0004592678034102307, "loss": 2.8427, "theoretical_loss": 4.147767568524845, "tokens_seen": 299171840 }, { "epoch": 0.09, "learning_rate": 0.0004592577733199599, "loss": 3.1985, "theoretical_loss": 4.147660927754942, "tokens_seen": 299237376 }, { "epoch": 0.09, "learning_rate": 0.0004592477432296891, "loss": 2.9435, "theoretical_loss": 4.147554316875766, "tokens_seen": 299302912 }, { "epoch": 0.09, "learning_rate": 0.00045923771313941827, "loss": 3.0539, "theoretical_loss": 4.147447735872396, "tokens_seen": 299368448 }, { "epoch": 0.09, "learning_rate": 0.00045922768304914745, "loss": 2.9897, "theoretical_loss": 4.147341184729921, "tokens_seen": 299433984 }, { "epoch": 0.09, "learning_rate": 0.00045921765295887663, "loss": 2.9482, "theoretical_loss": 4.147234663433444, "tokens_seen": 299499520 }, { "epoch": 0.09, "learning_rate": 0.00045920762286860586, "loss": 2.8237, "theoretical_loss": 4.147128171968077, "tokens_seen": 299565056 }, { "epoch": 0.09, "learning_rate": 0.000459197592778335, "loss": 3.1106, "theoretical_loss": 4.14702171031894, "tokens_seen": 299630592 }, { "epoch": 0.09, "learning_rate": 0.0004591875626880642, "loss": 2.9476, "theoretical_loss": 4.146915278471169, "tokens_seen": 299696128 }, { "epoch": 0.09, "learning_rate": 0.00045917753259779335, "loss": 2.968, "theoretical_loss": 4.146808876409906, "tokens_seen": 299761664 }, { "epoch": 0.09, "objective/train/docs_used": 181680, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3923745155334473, "objective/train/theoretical_loss": 4.146702504120305, "objective/train/tokens_used": 320287200, "theoretical_loss": 4.146702504120305, "tokens_seen": 299827200 }, { "epoch": 0.09, "learning_rate": 0.0004591675025075226, "loss": 2.9731, "theoretical_loss": 4.146702504120305, "tokens_seen": 299827200 }, { "epoch": 0.09, "learning_rate": 0.00045915747241725177, "loss": 2.7663, "theoretical_loss": 4.146596161587532, "tokens_seen": 299892736 }, { "epoch": 0.09, "learning_rate": 0.00045914744232698095, "loss": 2.5711, "theoretical_loss": 4.146489848796763, "tokens_seen": 299958272 }, { "epoch": 0.09, "learning_rate": 0.00045913741223671013, "loss": 3.2189, "theoretical_loss": 4.146383565733184, "tokens_seen": 300023808 }, { "epoch": 0.09, "learning_rate": 0.00045912738214643937, "loss": 3.0249, "theoretical_loss": 4.146277312381991, "tokens_seen": 300089344 }, { "epoch": 0.09, "learning_rate": 0.0004591173520561685, "loss": 2.903, "theoretical_loss": 4.1461710887283925, "tokens_seen": 300154880 }, { "epoch": 0.09, "learning_rate": 0.00045910732196589773, "loss": 3.1524, "theoretical_loss": 4.146064894757606, "tokens_seen": 300220416 }, { "epoch": 0.09, "learning_rate": 0.00045909729187562686, "loss": 2.9013, "theoretical_loss": 4.145958730454861, "tokens_seen": 300285952 }, { "epoch": 0.09, "learning_rate": 0.0004590872617853561, "loss": 2.9616, "theoretical_loss": 4.145852595805396, "tokens_seen": 300351488 }, { "epoch": 0.09, "learning_rate": 0.0004590772316950853, "loss": 3.0813, "theoretical_loss": 4.145746490794461, "tokens_seen": 300417024 }, { "epoch": 0.09, "learning_rate": 0.00045906720160481445, "loss": 3.1084, "theoretical_loss": 4.145640415407317, "tokens_seen": 300482560 }, { "epoch": 0.09, "learning_rate": 0.00045905717151454364, "loss": 2.9712, "theoretical_loss": 4.145534369629234, "tokens_seen": 300548096 }, { "epoch": 0.09, "learning_rate": 0.0004590471414242728, "loss": 2.9661, "theoretical_loss": 4.145428353445494, "tokens_seen": 300613632 }, { "epoch": 0.09, "learning_rate": 0.000459037111334002, "loss": 3.211, "theoretical_loss": 4.145322366841389, "tokens_seen": 300679168 }, { "epoch": 0.09, "learning_rate": 0.00045902708124373123, "loss": 2.9262, "theoretical_loss": 4.145216409802221, "tokens_seen": 300744704 }, { "epoch": 0.09, "learning_rate": 0.00045901705115346036, "loss": 2.9333, "theoretical_loss": 4.145110482313304, "tokens_seen": 300810240 }, { "epoch": 0.09, "learning_rate": 0.0004590070210631896, "loss": 3.0263, "theoretical_loss": 4.1450045843599606, "tokens_seen": 300875776 }, { "epoch": 0.09, "learning_rate": 0.0004589969909729187, "loss": 2.906, "theoretical_loss": 4.144898715927525, "tokens_seen": 300941312 }, { "epoch": 0.09, "learning_rate": 0.00045898696088264796, "loss": 2.8445, "theoretical_loss": 4.144792877001342, "tokens_seen": 301006848 }, { "epoch": 0.09, "learning_rate": 0.00045897693079237714, "loss": 2.984, "theoretical_loss": 4.144687067566765, "tokens_seen": 301072384 }, { "epoch": 0.09, "learning_rate": 0.0004589669007021063, "loss": 2.9933, "theoretical_loss": 4.144581287609161, "tokens_seen": 301137920 }, { "epoch": 0.09, "learning_rate": 0.0004589568706118355, "loss": 2.9471, "theoretical_loss": 4.144475537113905, "tokens_seen": 301203456 }, { "epoch": 0.09, "learning_rate": 0.00045894684052156474, "loss": 3.0003, "theoretical_loss": 4.144369816066385, "tokens_seen": 301268992 }, { "epoch": 0.09, "learning_rate": 0.00045893681043129386, "loss": 3.1696, "theoretical_loss": 4.144264124451995, "tokens_seen": 301334528 }, { "epoch": 0.09, "learning_rate": 0.0004589267803410231, "loss": 3.1541, "theoretical_loss": 4.1441584622561445, "tokens_seen": 301400064 }, { "epoch": 0.09, "objective/train/docs_used": 182385, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2453272342681885, "objective/train/theoretical_loss": 4.144052829464249, "objective/train/tokens_used": 321925600, "theoretical_loss": 4.144052829464249, "tokens_seen": 301465600 }, { "epoch": 0.09, "learning_rate": 0.0004589167502507522, "loss": 3.212, "theoretical_loss": 4.144052829464249, "tokens_seen": 301465600 }, { "epoch": 0.09, "learning_rate": 0.00045890672016048146, "loss": 3.221, "theoretical_loss": 4.143947226061737, "tokens_seen": 301531136 }, { "epoch": 0.09, "learning_rate": 0.00045889669007021064, "loss": 3.1414, "theoretical_loss": 4.143841652034048, "tokens_seen": 301596672 }, { "epoch": 0.09, "learning_rate": 0.0004588866599799398, "loss": 2.8214, "theoretical_loss": 4.143736107366629, "tokens_seen": 301662208 }, { "epoch": 0.09, "learning_rate": 0.000458876629889669, "loss": 2.9196, "theoretical_loss": 4.14363059204494, "tokens_seen": 301727744 }, { "epoch": 0.09, "learning_rate": 0.0004588665997993982, "loss": 2.8951, "theoretical_loss": 4.14352510605445, "tokens_seen": 301793280 }, { "epoch": 0.09, "learning_rate": 0.00045885656970912737, "loss": 3.2871, "theoretical_loss": 4.143419649380639, "tokens_seen": 301858816 }, { "epoch": 0.09, "learning_rate": 0.0004588465396188566, "loss": 3.0498, "theoretical_loss": 4.143314222008997, "tokens_seen": 301924352 }, { "epoch": 0.09, "learning_rate": 0.00045883650952858573, "loss": 3.2756, "theoretical_loss": 4.143208823925024, "tokens_seen": 301989888 }, { "epoch": 0.09, "learning_rate": 0.00045882647943831496, "loss": 2.7953, "theoretical_loss": 4.143103455114231, "tokens_seen": 302055424 }, { "epoch": 0.09, "learning_rate": 0.00045881644934804415, "loss": 2.9904, "theoretical_loss": 4.142998115562139, "tokens_seen": 302120960 }, { "epoch": 0.09, "learning_rate": 0.0004588064192577733, "loss": 2.8618, "theoretical_loss": 4.14289280525428, "tokens_seen": 302186496 }, { "epoch": 0.09, "learning_rate": 0.0004587963891675025, "loss": 2.8056, "theoretical_loss": 4.142787524176194, "tokens_seen": 302252032 }, { "epoch": 0.09, "learning_rate": 0.0004587863590772317, "loss": 2.9704, "theoretical_loss": 4.142682272313435, "tokens_seen": 302317568 }, { "epoch": 0.09, "learning_rate": 0.00045877632898696087, "loss": 3.2137, "theoretical_loss": 4.142577049651563, "tokens_seen": 302383104 }, { "epoch": 0.09, "learning_rate": 0.0004587662988966901, "loss": 3.2904, "theoretical_loss": 4.142471856176152, "tokens_seen": 302448640 }, { "epoch": 0.09, "learning_rate": 0.00045875626880641923, "loss": 2.9358, "theoretical_loss": 4.142366691872784, "tokens_seen": 302514176 }, { "epoch": 0.09, "learning_rate": 0.00045874623871614847, "loss": 3.1388, "theoretical_loss": 4.142261556727052, "tokens_seen": 302579712 }, { "epoch": 0.09, "learning_rate": 0.0004587362086258776, "loss": 2.8735, "theoretical_loss": 4.14215645072456, "tokens_seen": 302645248 }, { "epoch": 0.09, "learning_rate": 0.00045872617853560683, "loss": 2.9464, "theoretical_loss": 4.14205137385092, "tokens_seen": 302710784 }, { "epoch": 0.09, "learning_rate": 0.000458716148445336, "loss": 3.0534, "theoretical_loss": 4.141946326091756, "tokens_seen": 302776320 }, { "epoch": 0.09, "learning_rate": 0.0004587061183550652, "loss": 3.0235, "theoretical_loss": 4.141841307432703, "tokens_seen": 302841856 }, { "epoch": 0.09, "learning_rate": 0.0004586960882647944, "loss": 2.9913, "theoretical_loss": 4.1417363178594035, "tokens_seen": 302907392 }, { "epoch": 0.09, "learning_rate": 0.00045868605817452355, "loss": 3.1955, "theoretical_loss": 4.141631357357513, "tokens_seen": 302972928 }, { "epoch": 0.09, "learning_rate": 0.0004586760280842528, "loss": 3.1505, "theoretical_loss": 4.141526425912694, "tokens_seen": 303038464 }, { "epoch": 0.09, "objective/train/docs_used": 183646, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1540284156799316, "objective/train/theoretical_loss": 4.141421523510623, "objective/train/tokens_used": 323564000, "theoretical_loss": 4.141421523510623, "tokens_seen": 303104000 }, { "epoch": 0.09, "learning_rate": 0.00045866599799398197, "loss": 2.9251, "theoretical_loss": 4.141421523510623, "tokens_seen": 303104000 }, { "epoch": 0.09, "learning_rate": 0.00045865596790371115, "loss": 3.11, "theoretical_loss": 4.141316650136983, "tokens_seen": 303169536 }, { "epoch": 0.09, "learning_rate": 0.00045864593781344033, "loss": 2.9084, "theoretical_loss": 4.14121180577747, "tokens_seen": 303235072 }, { "epoch": 0.09, "learning_rate": 0.00045863590772316957, "loss": 3.2785, "theoretical_loss": 4.141106990417789, "tokens_seen": 303300608 }, { "epoch": 0.09, "learning_rate": 0.0004586258776328987, "loss": 2.9644, "theoretical_loss": 4.141002204043654, "tokens_seen": 303366144 }, { "epoch": 0.09, "learning_rate": 0.00045861584754262793, "loss": 3.293, "theoretical_loss": 4.140897446640793, "tokens_seen": 303431680 }, { "epoch": 0.09, "learning_rate": 0.00045860581745235706, "loss": 2.9656, "theoretical_loss": 4.1407927181949375, "tokens_seen": 303497216 }, { "epoch": 0.09, "learning_rate": 0.0004585957873620863, "loss": 2.9074, "theoretical_loss": 4.140688018691835, "tokens_seen": 303562752 }, { "epoch": 0.09, "learning_rate": 0.0004585857572718155, "loss": 2.884, "theoretical_loss": 4.140583348117241, "tokens_seen": 303628288 }, { "epoch": 0.09, "learning_rate": 0.00045857572718154465, "loss": 3.0019, "theoretical_loss": 4.140478706456921, "tokens_seen": 303693824 }, { "epoch": 0.09, "learning_rate": 0.00045856569709127384, "loss": 3.0583, "theoretical_loss": 4.140374093696651, "tokens_seen": 303759360 }, { "epoch": 0.09, "learning_rate": 0.000458555667001003, "loss": 3.1779, "theoretical_loss": 4.1402695098222155, "tokens_seen": 303824896 }, { "epoch": 0.09, "learning_rate": 0.0004585456369107322, "loss": 2.8, "theoretical_loss": 4.140164954819412, "tokens_seen": 303890432 }, { "epoch": 0.09, "learning_rate": 0.00045853560682046143, "loss": 3.1757, "theoretical_loss": 4.140060428674046, "tokens_seen": 303955968 }, { "epoch": 0.09, "learning_rate": 0.00045852557673019056, "loss": 3.1018, "theoretical_loss": 4.139955931371932, "tokens_seen": 304021504 }, { "epoch": 0.09, "learning_rate": 0.0004585155466399198, "loss": 2.8922, "theoretical_loss": 4.139851462898897, "tokens_seen": 304087040 }, { "epoch": 0.09, "learning_rate": 0.0004585055165496489, "loss": 3.0495, "theoretical_loss": 4.139747023240777, "tokens_seen": 304152576 }, { "epoch": 0.09, "learning_rate": 0.00045849548645937816, "loss": 3.159, "theoretical_loss": 4.139642612383418, "tokens_seen": 304218112 }, { "epoch": 0.09, "learning_rate": 0.00045848545636910734, "loss": 3.3414, "theoretical_loss": 4.1395382303126755, "tokens_seen": 304283648 }, { "epoch": 0.09, "learning_rate": 0.0004584754262788365, "loss": 3.0128, "theoretical_loss": 4.139433877014415, "tokens_seen": 304349184 }, { "epoch": 0.09, "learning_rate": 0.0004584653961885657, "loss": 3.2096, "theoretical_loss": 4.139329552474514, "tokens_seen": 304414720 }, { "epoch": 0.09, "learning_rate": 0.00045845536609829494, "loss": 2.9645, "theoretical_loss": 4.139225256678857, "tokens_seen": 304480256 }, { "epoch": 0.09, "learning_rate": 0.00045844533600802406, "loss": 3.2456, "theoretical_loss": 4.139120989613341, "tokens_seen": 304545792 }, { "epoch": 0.09, "learning_rate": 0.0004584353059177533, "loss": 3.2019, "theoretical_loss": 4.1390167512638705, "tokens_seen": 304611328 }, { "epoch": 0.09, "learning_rate": 0.0004584252758274824, "loss": 3.2112, "theoretical_loss": 4.138912541616363, "tokens_seen": 304676864 }, { "epoch": 0.09, "objective/train/docs_used": 184075, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.958650827407837, "objective/train/theoretical_loss": 4.138808360656742, "objective/train/tokens_used": 325202400, "theoretical_loss": 4.138808360656742, "tokens_seen": 304742400 }, { "epoch": 0.09, "learning_rate": 0.00045841524573721166, "loss": 2.9379, "theoretical_loss": 4.138808360656742, "tokens_seen": 304742400 }, { "epoch": 0.09, "learning_rate": 0.00045840521564694084, "loss": 3.0492, "theoretical_loss": 4.138704208370944, "tokens_seen": 304807936 }, { "epoch": 0.09, "learning_rate": 0.00045839518555667, "loss": 2.9333, "theoretical_loss": 4.138600084744915, "tokens_seen": 304873472 }, { "epoch": 0.09, "learning_rate": 0.0004583851554663992, "loss": 3.0129, "theoretical_loss": 4.1384959897646105, "tokens_seen": 304939008 }, { "epoch": 0.09, "learning_rate": 0.0004583751253761284, "loss": 3.1081, "theoretical_loss": 4.138391923415996, "tokens_seen": 305004544 }, { "epoch": 0.09, "learning_rate": 0.00045836509528585757, "loss": 3.0953, "theoretical_loss": 4.138287885685045, "tokens_seen": 305070080 }, { "epoch": 0.09, "learning_rate": 0.0004583550651955868, "loss": 3.0357, "theoretical_loss": 4.138183876557745, "tokens_seen": 305135616 }, { "epoch": 0.09, "learning_rate": 0.00045834503510531593, "loss": 3.2149, "theoretical_loss": 4.1380798960200895, "tokens_seen": 305201152 }, { "epoch": 0.09, "learning_rate": 0.00045833500501504516, "loss": 3.0338, "theoretical_loss": 4.137975944058083, "tokens_seen": 305266688 }, { "epoch": 0.09, "learning_rate": 0.00045832497492477435, "loss": 3.0291, "theoretical_loss": 4.137872020657742, "tokens_seen": 305332224 }, { "epoch": 0.09, "learning_rate": 0.0004583149448345035, "loss": 3.1508, "theoretical_loss": 4.1377681258050885, "tokens_seen": 305397760 }, { "epoch": 0.09, "learning_rate": 0.0004583049147442327, "loss": 2.9683, "theoretical_loss": 4.13766425948616, "tokens_seen": 305463296 }, { "epoch": 0.09, "learning_rate": 0.0004582948846539619, "loss": 3.0612, "theoretical_loss": 4.137560421686998, "tokens_seen": 305528832 }, { "epoch": 0.09, "learning_rate": 0.00045828485456369107, "loss": 2.8031, "theoretical_loss": 4.137456612393658, "tokens_seen": 305594368 }, { "epoch": 0.09, "learning_rate": 0.0004582748244734203, "loss": 3.2981, "theoretical_loss": 4.137352831592203, "tokens_seen": 305659904 }, { "epoch": 0.09, "learning_rate": 0.00045826479438314943, "loss": 3.1212, "theoretical_loss": 4.137249079268707, "tokens_seen": 305725440 }, { "epoch": 0.09, "learning_rate": 0.00045825476429287867, "loss": 3.1303, "theoretical_loss": 4.137145355409253, "tokens_seen": 305790976 }, { "epoch": 0.09, "learning_rate": 0.0004582447342026078, "loss": 3.0405, "theoretical_loss": 4.137041659999936, "tokens_seen": 305856512 }, { "epoch": 0.09, "learning_rate": 0.00045823470411233703, "loss": 3.0228, "theoretical_loss": 4.136937993026857, "tokens_seen": 305922048 }, { "epoch": 0.09, "learning_rate": 0.0004582246740220662, "loss": 3.0352, "theoretical_loss": 4.136834354476129, "tokens_seen": 305987584 }, { "epoch": 0.09, "learning_rate": 0.0004582146439317954, "loss": 3.0844, "theoretical_loss": 4.1367307443338746, "tokens_seen": 306053120 }, { "epoch": 0.09, "learning_rate": 0.00045820461384152457, "loss": 3.0252, "theoretical_loss": 4.136627162586226, "tokens_seen": 306118656 }, { "epoch": 0.09, "learning_rate": 0.00045819458375125375, "loss": 2.9533, "theoretical_loss": 4.136523609219327, "tokens_seen": 306184192 }, { "epoch": 0.09, "learning_rate": 0.00045818455366098294, "loss": 3.0824, "theoretical_loss": 4.136420084219327, "tokens_seen": 306249728 }, { "epoch": 0.09, "learning_rate": 0.00045817452357071217, "loss": 2.8966, "theoretical_loss": 4.136316587572388, "tokens_seen": 306315264 }, { "epoch": 0.09, "objective/train/docs_used": 185377, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.6815130710601807, "objective/train/theoretical_loss": 4.136213119264681, "objective/train/tokens_used": 326840800, "theoretical_loss": 4.136213119264681, "tokens_seen": 306380800 }, { "epoch": 0.09, "learning_rate": 0.0004581644934804413, "loss": 3.2741, "theoretical_loss": 4.136213119264681, "tokens_seen": 306380800 }, { "epoch": 0.09, "learning_rate": 0.00045815446339017053, "loss": 3.0248, "theoretical_loss": 4.136109679282388, "tokens_seen": 306446336 }, { "epoch": 0.09, "learning_rate": 0.0004581444332998997, "loss": 3.086, "theoretical_loss": 4.136006267611697, "tokens_seen": 306511872 }, { "epoch": 0.09, "learning_rate": 0.0004581344032096289, "loss": 3.1913, "theoretical_loss": 4.135902884238812, "tokens_seen": 306577408 }, { "epoch": 0.09, "learning_rate": 0.0004581243731193581, "loss": 3.2079, "theoretical_loss": 4.135799529149939, "tokens_seen": 306642944 }, { "epoch": 0.09, "learning_rate": 0.00045811434302908726, "loss": 3.0883, "theoretical_loss": 4.1356962023313, "tokens_seen": 306708480 }, { "epoch": 0.09, "learning_rate": 0.00045810431293881644, "loss": 3.0245, "theoretical_loss": 4.135592903769124, "tokens_seen": 306774016 }, { "epoch": 0.09, "learning_rate": 0.0004580942828485457, "loss": 2.9307, "theoretical_loss": 4.135489633449649, "tokens_seen": 306839552 }, { "epoch": 0.09, "learning_rate": 0.0004580842527582748, "loss": 2.8345, "theoretical_loss": 4.135386391359123, "tokens_seen": 306905088 }, { "epoch": 0.09, "learning_rate": 0.00045807422266800404, "loss": 3.2493, "theoretical_loss": 4.135283177483807, "tokens_seen": 306970624 }, { "epoch": 0.09, "learning_rate": 0.00045806419257773316, "loss": 3.0784, "theoretical_loss": 4.135179991809965, "tokens_seen": 307036160 }, { "epoch": 0.09, "learning_rate": 0.0004580541624874624, "loss": 2.9564, "theoretical_loss": 4.135076834323876, "tokens_seen": 307101696 }, { "epoch": 0.09, "learning_rate": 0.0004580441323971916, "loss": 3.2303, "theoretical_loss": 4.134973705011828, "tokens_seen": 307167232 }, { "epoch": 0.09, "learning_rate": 0.00045803410230692076, "loss": 2.8665, "theoretical_loss": 4.134870603860117, "tokens_seen": 307232768 }, { "epoch": 0.09, "learning_rate": 0.00045802407221664994, "loss": 3.1065, "theoretical_loss": 4.134767530855047, "tokens_seen": 307298304 }, { "epoch": 0.09, "learning_rate": 0.0004580140421263791, "loss": 3.1849, "theoretical_loss": 4.1346644859829365, "tokens_seen": 307363840 }, { "epoch": 0.09, "learning_rate": 0.0004580040120361083, "loss": 3.1406, "theoretical_loss": 4.1345614692301105, "tokens_seen": 307429376 }, { "epoch": 0.09, "learning_rate": 0.00045799398194583754, "loss": 3.0077, "theoretical_loss": 4.134458480582902, "tokens_seen": 307494912 }, { "epoch": 0.09, "learning_rate": 0.00045798395185556667, "loss": 2.9305, "theoretical_loss": 4.134355520027657, "tokens_seen": 307560448 }, { "epoch": 0.09, "learning_rate": 0.0004579739217652959, "loss": 3.3013, "theoretical_loss": 4.134252587550728, "tokens_seen": 307625984 }, { "epoch": 0.09, "learning_rate": 0.0004579638916750251, "loss": 2.8856, "theoretical_loss": 4.134149683138481, "tokens_seen": 307691520 }, { "epoch": 0.09, "learning_rate": 0.00045795386158475426, "loss": 2.9128, "theoretical_loss": 4.134046806777286, "tokens_seen": 307757056 }, { "epoch": 0.09, "learning_rate": 0.0004579438314944835, "loss": 3.0849, "theoretical_loss": 4.133943958453528, "tokens_seen": 307822592 }, { "epoch": 0.09, "learning_rate": 0.0004579338014042126, "loss": 2.9414, "theoretical_loss": 4.133841138153597, "tokens_seen": 307888128 }, { "epoch": 0.09, "learning_rate": 0.00045792377131394186, "loss": 3.2586, "theoretical_loss": 4.133738345863896, "tokens_seen": 307953664 }, { "epoch": 0.09, "objective/train/docs_used": 186163, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.693579912185669, "objective/train/theoretical_loss": 4.133635581570836, "objective/train/tokens_used": 328479200, "theoretical_loss": 4.133635581570836, "tokens_seen": 308019200 }, { "epoch": 0.09, "learning_rate": 0.00045791374122367104, "loss": 3.0399, "theoretical_loss": 4.133635581570836, "tokens_seen": 308019200 }, { "epoch": 0.09, "learning_rate": 0.0004579037111334002, "loss": 3.0709, "theoretical_loss": 4.133532845260836, "tokens_seen": 308084736 }, { "epoch": 0.09, "learning_rate": 0.0004578936810431294, "loss": 3.0057, "theoretical_loss": 4.133430136920327, "tokens_seen": 308150272 }, { "epoch": 0.09, "learning_rate": 0.0004578836509528586, "loss": 3.1167, "theoretical_loss": 4.133327456535749, "tokens_seen": 308215808 }, { "epoch": 0.09, "learning_rate": 0.00045787362086258777, "loss": 2.8566, "theoretical_loss": 4.13322480409355, "tokens_seen": 308281344 }, { "epoch": 0.09, "learning_rate": 0.000457863590772317, "loss": 2.9801, "theoretical_loss": 4.133122179580189, "tokens_seen": 308346880 }, { "epoch": 0.09, "learning_rate": 0.00045785356068204613, "loss": 2.9788, "theoretical_loss": 4.133019582982134, "tokens_seen": 308412416 }, { "epoch": 0.09, "learning_rate": 0.00045784353059177536, "loss": 3.0067, "theoretical_loss": 4.1329170142858604, "tokens_seen": 308477952 }, { "epoch": 0.09, "learning_rate": 0.00045783350050150455, "loss": 3.0007, "theoretical_loss": 4.132814473477857, "tokens_seen": 308543488 }, { "epoch": 0.09, "learning_rate": 0.0004578234704112337, "loss": 3.2335, "theoretical_loss": 4.1327119605446185, "tokens_seen": 308609024 }, { "epoch": 0.09, "learning_rate": 0.0004578134403209629, "loss": 2.903, "theoretical_loss": 4.132609475472651, "tokens_seen": 308674560 }, { "epoch": 0.09, "learning_rate": 0.0004578034102306921, "loss": 3.0954, "theoretical_loss": 4.132507018248469, "tokens_seen": 308740096 }, { "epoch": 0.09, "learning_rate": 0.00045779338014042127, "loss": 3.0897, "theoretical_loss": 4.132404588858597, "tokens_seen": 308805632 }, { "epoch": 0.09, "learning_rate": 0.0004577833500501505, "loss": 3.0264, "theoretical_loss": 4.132302187289568, "tokens_seen": 308871168 }, { "epoch": 0.09, "learning_rate": 0.00045777331995987963, "loss": 3.3141, "theoretical_loss": 4.132199813527926, "tokens_seen": 308936704 }, { "epoch": 0.09, "learning_rate": 0.00045776328986960887, "loss": 3.1779, "theoretical_loss": 4.132097467560223, "tokens_seen": 309002240 }, { "epoch": 0.09, "learning_rate": 0.000457753259779338, "loss": 3.2129, "theoretical_loss": 4.1319951493730205, "tokens_seen": 309067776 }, { "epoch": 0.09, "learning_rate": 0.00045774322968906723, "loss": 3.3985, "theoretical_loss": 4.131892858952889, "tokens_seen": 309133312 }, { "epoch": 0.09, "learning_rate": 0.0004577331995987964, "loss": 3.0588, "theoretical_loss": 4.131790596286409, "tokens_seen": 309198848 }, { "epoch": 0.09, "learning_rate": 0.0004577231695085256, "loss": 2.8701, "theoretical_loss": 4.1316883613601725, "tokens_seen": 309264384 }, { "epoch": 0.09, "learning_rate": 0.0004577131394182548, "loss": 3.0681, "theoretical_loss": 4.131586154160775, "tokens_seen": 309329920 }, { "epoch": 0.09, "learning_rate": 0.00045770310932798395, "loss": 3.0305, "theoretical_loss": 4.131483974674827, "tokens_seen": 309395456 }, { "epoch": 0.09, "learning_rate": 0.00045769307923771314, "loss": 2.9912, "theoretical_loss": 4.131381822888946, "tokens_seen": 309460992 }, { "epoch": 0.09, "learning_rate": 0.00045768304914744237, "loss": 2.9485, "theoretical_loss": 4.131279698789759, "tokens_seen": 309526528 }, { "epoch": 0.09, "learning_rate": 0.0004576730190571715, "loss": 3.1318, "theoretical_loss": 4.1311776023639, "tokens_seen": 309592064 }, { "epoch": 0.09, "objective/train/docs_used": 187126, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0981154441833496, "objective/train/theoretical_loss": 4.131075533598018, "objective/train/tokens_used": 330117600, "theoretical_loss": 4.131075533598018, "tokens_seen": 309657600 }, { "epoch": 0.09, "learning_rate": 0.00045766298896690073, "loss": 3.0551, "theoretical_loss": 4.131075533598018, "tokens_seen": 309657600 }, { "epoch": 0.09, "learning_rate": 0.0004576529588766299, "loss": 3.3009, "theoretical_loss": 4.130973492478766, "tokens_seen": 309723136 }, { "epoch": 0.09, "learning_rate": 0.0004576429287863591, "loss": 3.293, "theoretical_loss": 4.130871478992807, "tokens_seen": 309788672 }, { "epoch": 0.09, "learning_rate": 0.0004576328986960883, "loss": 3.2951, "theoretical_loss": 4.130769493126817, "tokens_seen": 309854208 }, { "epoch": 0.09, "learning_rate": 0.00045762286860581746, "loss": 3.1751, "theoretical_loss": 4.130667534867476, "tokens_seen": 309919744 }, { "epoch": 0.09, "learning_rate": 0.00045761283851554664, "loss": 3.2966, "theoretical_loss": 4.130565604201477, "tokens_seen": 309985280 }, { "epoch": 0.09, "learning_rate": 0.0004576028084252759, "loss": 2.9467, "theoretical_loss": 4.130463701115521, "tokens_seen": 310050816 }, { "epoch": 0.09, "learning_rate": 0.000457592778335005, "loss": 2.7801, "theoretical_loss": 4.130361825596317, "tokens_seen": 310116352 }, { "epoch": 0.09, "learning_rate": 0.00045758274824473424, "loss": 3.1128, "theoretical_loss": 4.130259977630586, "tokens_seen": 310181888 }, { "epoch": 0.09, "learning_rate": 0.00045757271815446336, "loss": 3.0139, "theoretical_loss": 4.130158157205056, "tokens_seen": 310247424 }, { "epoch": 0.09, "learning_rate": 0.0004575626880641926, "loss": 3.1759, "theoretical_loss": 4.130056364306465, "tokens_seen": 310312960 }, { "epoch": 0.09, "learning_rate": 0.0004575526579739218, "loss": 3.0092, "theoretical_loss": 4.129954598921559, "tokens_seen": 310378496 }, { "epoch": 0.09, "learning_rate": 0.00045754262788365096, "loss": 2.7612, "theoretical_loss": 4.1298528610370955, "tokens_seen": 310444032 }, { "epoch": 0.09, "learning_rate": 0.00045753259779338014, "loss": 3.0608, "theoretical_loss": 4.12975115063984, "tokens_seen": 310509568 }, { "epoch": 0.09, "learning_rate": 0.0004575225677031093, "loss": 2.6856, "theoretical_loss": 4.129649467716565, "tokens_seen": 310575104 }, { "epoch": 0.09, "learning_rate": 0.0004575125376128385, "loss": 3.2868, "theoretical_loss": 4.1295478122540565, "tokens_seen": 310640640 }, { "epoch": 0.09, "learning_rate": 0.00045750250752256774, "loss": 3.2094, "theoretical_loss": 4.1294461842391055, "tokens_seen": 310706176 }, { "epoch": 0.09, "learning_rate": 0.00045749247743229687, "loss": 2.9271, "theoretical_loss": 4.129344583658516, "tokens_seen": 310771712 }, { "epoch": 0.09, "learning_rate": 0.0004574824473420261, "loss": 3.1971, "theoretical_loss": 4.1292430104990965, "tokens_seen": 310837248 }, { "epoch": 0.09, "learning_rate": 0.0004574724172517553, "loss": 2.8271, "theoretical_loss": 4.1291414647476685, "tokens_seen": 310902784 }, { "epoch": 0.09, "learning_rate": 0.00045746238716148446, "loss": 2.9373, "theoretical_loss": 4.129039946391062, "tokens_seen": 310968320 }, { "epoch": 0.09, "learning_rate": 0.00045745235707121364, "loss": 3.2863, "theoretical_loss": 4.128938455416115, "tokens_seen": 311033856 }, { "epoch": 0.09, "learning_rate": 0.0004574423269809428, "loss": 3.0255, "theoretical_loss": 4.128836991809674, "tokens_seen": 311099392 }, { "epoch": 0.09, "learning_rate": 0.000457432296890672, "loss": 3.1146, "theoretical_loss": 4.128735555558597, "tokens_seen": 311164928 }, { "epoch": 0.09, "learning_rate": 0.00045742226680040124, "loss": 2.9283, "theoretical_loss": 4.128634146649748, "tokens_seen": 311230464 }, { "epoch": 0.09, "objective/train/docs_used": 187875, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4823386669158936, "objective/train/theoretical_loss": 4.128532765070004, "objective/train/tokens_used": 331756000, "theoretical_loss": 4.128532765070004, "tokens_seen": 311296000 }, { "epoch": 0.09, "learning_rate": 0.00045741223671013037, "loss": 3.1208, "theoretical_loss": 4.128532765070004, "tokens_seen": 311296000 }, { "epoch": 0.09, "learning_rate": 0.0004574022066198596, "loss": 2.94, "theoretical_loss": 4.128431410806247, "tokens_seen": 311361536 }, { "epoch": 0.09, "learning_rate": 0.00045739217652958873, "loss": 2.9579, "theoretical_loss": 4.12833008384537, "tokens_seen": 311427072 }, { "epoch": 0.09, "learning_rate": 0.00045738214643931797, "loss": 3.2517, "theoretical_loss": 4.128228784174275, "tokens_seen": 311492608 }, { "epoch": 0.09, "learning_rate": 0.00045737211634904715, "loss": 3.2745, "theoretical_loss": 4.128127511779873, "tokens_seen": 311558144 }, { "epoch": 0.09, "learning_rate": 0.00045736208625877633, "loss": 2.9142, "theoretical_loss": 4.128026266649085, "tokens_seen": 311623680 }, { "epoch": 0.09, "learning_rate": 0.0004573520561685055, "loss": 3.03, "theoretical_loss": 4.127925048768839, "tokens_seen": 311689216 }, { "epoch": 0.09, "learning_rate": 0.00045734202607823475, "loss": 3.068, "theoretical_loss": 4.127823858126073, "tokens_seen": 311754752 }, { "epoch": 0.09, "learning_rate": 0.00045733199598796387, "loss": 3.1706, "theoretical_loss": 4.1277226947077335, "tokens_seen": 311820288 }, { "epoch": 0.09, "learning_rate": 0.0004573219658976931, "loss": 2.9845, "theoretical_loss": 4.127621558500778, "tokens_seen": 311885824 }, { "epoch": 0.09, "learning_rate": 0.00045731193580742223, "loss": 2.9558, "theoretical_loss": 4.12752044949217, "tokens_seen": 311951360 }, { "epoch": 0.09, "learning_rate": 0.00045730190571715147, "loss": 3.0432, "theoretical_loss": 4.127419367668884, "tokens_seen": 312016896 }, { "epoch": 0.09, "learning_rate": 0.00045729187562688065, "loss": 2.9326, "theoretical_loss": 4.127318313017904, "tokens_seen": 312082432 }, { "epoch": 0.09, "learning_rate": 0.00045728184553660983, "loss": 3.057, "theoretical_loss": 4.12721728552622, "tokens_seen": 312147968 }, { "epoch": 0.09, "learning_rate": 0.000457271815446339, "loss": 2.9781, "theoretical_loss": 4.1271162851808345, "tokens_seen": 312213504 }, { "epoch": 0.09, "learning_rate": 0.0004572617853560682, "loss": 3.2397, "theoretical_loss": 4.127015311968757, "tokens_seen": 312279040 }, { "epoch": 0.09, "learning_rate": 0.0004572517552657974, "loss": 3.0724, "theoretical_loss": 4.126914365877004, "tokens_seen": 312344576 }, { "epoch": 0.09, "learning_rate": 0.0004572417251755266, "loss": 3.1184, "theoretical_loss": 4.126813446892607, "tokens_seen": 312410112 }, { "epoch": 0.09, "learning_rate": 0.00045723169508525574, "loss": 3.0199, "theoretical_loss": 4.1267125550026, "tokens_seen": 312475648 }, { "epoch": 0.09, "learning_rate": 0.000457221664994985, "loss": 3.1512, "theoretical_loss": 4.1266116901940295, "tokens_seen": 312541184 }, { "epoch": 0.09, "learning_rate": 0.0004572116349047141, "loss": 2.9637, "theoretical_loss": 4.126510852453949, "tokens_seen": 312606720 }, { "epoch": 0.09, "learning_rate": 0.00045720160481444334, "loss": 3.1577, "theoretical_loss": 4.126410041769423, "tokens_seen": 312672256 }, { "epoch": 0.09, "learning_rate": 0.00045719157472417257, "loss": 2.8805, "theoretical_loss": 4.126309258127524, "tokens_seen": 312737792 }, { "epoch": 0.09, "learning_rate": 0.0004571815446339017, "loss": 3.2546, "theoretical_loss": 4.126208501515331, "tokens_seen": 312803328 }, { "epoch": 0.09, "learning_rate": 0.00045717151454363093, "loss": 3.0705, "theoretical_loss": 4.126107771919935, "tokens_seen": 312868864 }, { "epoch": 0.09, "objective/train/docs_used": 188927, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9257431030273438, "objective/train/theoretical_loss": 4.126007069328436, "objective/train/tokens_used": 333394400, "theoretical_loss": 4.126007069328436, "tokens_seen": 312934400 }, { "epoch": 0.09, "learning_rate": 0.0004571614844533601, "loss": 3.1474, "theoretical_loss": 4.126007069328436, "tokens_seen": 312934400 }, { "epoch": 0.09, "learning_rate": 0.0004571514543630893, "loss": 3.0964, "theoretical_loss": 4.125906393727941, "tokens_seen": 312999936 }, { "epoch": 0.09, "learning_rate": 0.0004571414242728185, "loss": 3.2524, "theoretical_loss": 4.125805745105566, "tokens_seen": 313065472 }, { "epoch": 0.09, "learning_rate": 0.00045713139418254766, "loss": 3.0474, "theoretical_loss": 4.125705123448437, "tokens_seen": 313131008 }, { "epoch": 0.09, "learning_rate": 0.00045712136409227684, "loss": 3.2728, "theoretical_loss": 4.125604528743689, "tokens_seen": 313196544 }, { "epoch": 0.09, "learning_rate": 0.0004571113340020061, "loss": 3.2944, "theoretical_loss": 4.125503960978464, "tokens_seen": 313262080 }, { "epoch": 0.09, "learning_rate": 0.0004571013039117352, "loss": 3.0976, "theoretical_loss": 4.1254034201399135, "tokens_seen": 313327616 }, { "epoch": 0.09, "learning_rate": 0.00045709127382146444, "loss": 3.0754, "theoretical_loss": 4.125302906215199, "tokens_seen": 313393152 }, { "epoch": 0.09, "learning_rate": 0.00045708124373119356, "loss": 3.1683, "theoretical_loss": 4.12520241919149, "tokens_seen": 313458688 }, { "epoch": 0.1, "learning_rate": 0.0004570712136409228, "loss": 3.1685, "theoretical_loss": 4.125101959055965, "tokens_seen": 313524224 }, { "epoch": 0.1, "learning_rate": 0.000457061183550652, "loss": 3.3666, "theoretical_loss": 4.125001525795811, "tokens_seen": 313589760 }, { "epoch": 0.1, "learning_rate": 0.00045705115346038116, "loss": 3.0122, "theoretical_loss": 4.124901119398222, "tokens_seen": 313655296 }, { "epoch": 0.1, "learning_rate": 0.00045704112337011034, "loss": 3.522, "theoretical_loss": 4.124800739850406, "tokens_seen": 313720832 }, { "epoch": 0.1, "learning_rate": 0.0004570310932798395, "loss": 2.9341, "theoretical_loss": 4.124700387139574, "tokens_seen": 313786368 }, { "epoch": 0.1, "learning_rate": 0.0004570210631895687, "loss": 3.1314, "theoretical_loss": 4.12460006125295, "tokens_seen": 313851904 }, { "epoch": 0.1, "learning_rate": 0.00045701103309929794, "loss": 3.1766, "theoretical_loss": 4.124499762177764, "tokens_seen": 313917440 }, { "epoch": 0.1, "learning_rate": 0.00045700100300902707, "loss": 3.2033, "theoretical_loss": 4.124399489901254, "tokens_seen": 313982976 }, { "epoch": 0.1, "learning_rate": 0.0004569909729187563, "loss": 3.1851, "theoretical_loss": 4.124299244410672, "tokens_seen": 314048512 }, { "epoch": 0.1, "learning_rate": 0.0004569809428284855, "loss": 3.3205, "theoretical_loss": 4.124199025693272, "tokens_seen": 314114048 }, { "epoch": 0.1, "learning_rate": 0.00045697091273821466, "loss": 2.9844, "theoretical_loss": 4.124098833736321, "tokens_seen": 314179584 }, { "epoch": 0.1, "learning_rate": 0.00045696088264794384, "loss": 3.2222, "theoretical_loss": 4.123998668527094, "tokens_seen": 314245120 }, { "epoch": 0.1, "learning_rate": 0.000456950852557673, "loss": 3.1528, "theoretical_loss": 4.123898530052874, "tokens_seen": 314310656 }, { "epoch": 0.1, "learning_rate": 0.0004569408224674022, "loss": 3.2758, "theoretical_loss": 4.123798418300953, "tokens_seen": 314376192 }, { "epoch": 0.1, "learning_rate": 0.00045693079237713144, "loss": 3.1248, "theoretical_loss": 4.123698333258631, "tokens_seen": 314441728 }, { "epoch": 0.1, "learning_rate": 0.00045692076228686057, "loss": 3.2705, "theoretical_loss": 4.123598274913219, "tokens_seen": 314507264 }, { "epoch": 0.1, "objective/train/docs_used": 189547, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8021438121795654, "objective/train/theoretical_loss": 4.123498243252032, "objective/train/tokens_used": 335032800, "theoretical_loss": 4.123498243252032, "tokens_seen": 314572800 }, { "epoch": 0.1, "learning_rate": 0.0004569107321965898, "loss": 2.9524, "theoretical_loss": 4.123498243252032, "tokens_seen": 314572800 }, { "epoch": 0.1, "learning_rate": 0.00045690070210631893, "loss": 2.9818, "theoretical_loss": 4.1233982382624, "tokens_seen": 314638336 }, { "epoch": 0.1, "learning_rate": 0.00045689067201604817, "loss": 3.1294, "theoretical_loss": 4.123298259931657, "tokens_seen": 314703872 }, { "epoch": 0.1, "learning_rate": 0.00045688064192577735, "loss": 3.2011, "theoretical_loss": 4.123198308247146, "tokens_seen": 314769408 }, { "epoch": 0.1, "learning_rate": 0.00045687061183550653, "loss": 3.2102, "theoretical_loss": 4.123098383196222, "tokens_seen": 314834944 }, { "epoch": 0.1, "learning_rate": 0.0004568605817452357, "loss": 3.0984, "theoretical_loss": 4.122998484766244, "tokens_seen": 314900480 }, { "epoch": 0.1, "learning_rate": 0.00045685055165496495, "loss": 3.0736, "theoretical_loss": 4.122898612944582, "tokens_seen": 314966016 }, { "epoch": 0.1, "learning_rate": 0.00045684052156469407, "loss": 3.4115, "theoretical_loss": 4.122798767718616, "tokens_seen": 315031552 }, { "epoch": 0.1, "learning_rate": 0.0004568304914744233, "loss": 3.2703, "theoretical_loss": 4.122698949075732, "tokens_seen": 315097088 }, { "epoch": 0.1, "learning_rate": 0.00045682046138415243, "loss": 3.3034, "theoretical_loss": 4.122599157003327, "tokens_seen": 315162624 }, { "epoch": 0.1, "learning_rate": 0.00045681043129388167, "loss": 3.0698, "theoretical_loss": 4.1224993914888035, "tokens_seen": 315228160 }, { "epoch": 0.1, "learning_rate": 0.00045680040120361085, "loss": 3.2926, "theoretical_loss": 4.122399652519576, "tokens_seen": 315293696 }, { "epoch": 0.1, "learning_rate": 0.00045679037111334003, "loss": 3.0107, "theoretical_loss": 4.122299940083065, "tokens_seen": 315359232 }, { "epoch": 0.1, "learning_rate": 0.0004567803410230692, "loss": 3.1404, "theoretical_loss": 4.1222002541667, "tokens_seen": 315424768 }, { "epoch": 0.1, "learning_rate": 0.0004567703109327984, "loss": 3.2998, "theoretical_loss": 4.122100594757921, "tokens_seen": 315490304 }, { "epoch": 0.1, "learning_rate": 0.0004567602808425276, "loss": 3.2533, "theoretical_loss": 4.122000961844175, "tokens_seen": 315555840 }, { "epoch": 0.1, "learning_rate": 0.0004567502507522568, "loss": 3.2643, "theoretical_loss": 4.121901355412917, "tokens_seen": 315621376 }, { "epoch": 0.1, "learning_rate": 0.00045674022066198594, "loss": 3.0887, "theoretical_loss": 4.121801775451612, "tokens_seen": 315686912 }, { "epoch": 0.1, "learning_rate": 0.0004567301905717152, "loss": 3.0258, "theoretical_loss": 4.121702221947732, "tokens_seen": 315752448 }, { "epoch": 0.1, "learning_rate": 0.0004567201604814443, "loss": 3.0984, "theoretical_loss": 4.121602694888759, "tokens_seen": 315817984 }, { "epoch": 0.1, "learning_rate": 0.00045671013039117354, "loss": 3.0961, "theoretical_loss": 4.121503194262183, "tokens_seen": 315883520 }, { "epoch": 0.1, "learning_rate": 0.0004567001003009027, "loss": 3.4246, "theoretical_loss": 4.121403720055502, "tokens_seen": 315949056 }, { "epoch": 0.1, "learning_rate": 0.0004566900702106319, "loss": 3.1232, "theoretical_loss": 4.121304272256222, "tokens_seen": 316014592 }, { "epoch": 0.1, "learning_rate": 0.0004566800401203611, "loss": 3.2672, "theoretical_loss": 4.121204850851861, "tokens_seen": 316080128 }, { "epoch": 0.1, "learning_rate": 0.0004566700100300903, "loss": 3.2468, "theoretical_loss": 4.121105455829939, "tokens_seen": 316145664 }, { "epoch": 0.1, "objective/train/docs_used": 190733, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.28241229057312, "objective/train/theoretical_loss": 4.121006087177992, "objective/train/tokens_used": 336671200, "theoretical_loss": 4.121006087177992, "tokens_seen": 316211200 }, { "epoch": 0.1, "learning_rate": 0.00045665997993981944, "loss": 3.0047, "theoretical_loss": 4.121006087177992, "tokens_seen": 316211200 }, { "epoch": 0.1, "learning_rate": 0.0004566499498495487, "loss": 3.0604, "theoretical_loss": 4.120906744883559, "tokens_seen": 316276736 }, { "epoch": 0.1, "learning_rate": 0.0004566399197592778, "loss": 3.1629, "theoretical_loss": 4.120807428934189, "tokens_seen": 316342272 }, { "epoch": 0.1, "learning_rate": 0.00045662988966900704, "loss": 3.0635, "theoretical_loss": 4.120708139317441, "tokens_seen": 316407808 }, { "epoch": 0.1, "learning_rate": 0.0004566198595787362, "loss": 3.2192, "theoretical_loss": 4.12060887602088, "tokens_seen": 316473344 }, { "epoch": 0.1, "learning_rate": 0.0004566098294884654, "loss": 3.1621, "theoretical_loss": 4.120509639032081, "tokens_seen": 316538880 }, { "epoch": 0.1, "learning_rate": 0.0004565997993981946, "loss": 2.9827, "theoretical_loss": 4.120410428338628, "tokens_seen": 316604416 }, { "epoch": 0.1, "learning_rate": 0.00045658976930792376, "loss": 2.9799, "theoretical_loss": 4.120311243928111, "tokens_seen": 316669952 }, { "epoch": 0.1, "learning_rate": 0.00045657973921765294, "loss": 3.195, "theoretical_loss": 4.120212085788131, "tokens_seen": 316735488 }, { "epoch": 0.1, "learning_rate": 0.0004565697091273822, "loss": 3.0131, "theoretical_loss": 4.120112953906296, "tokens_seen": 316801024 }, { "epoch": 0.1, "learning_rate": 0.0004565596790371113, "loss": 3.2145, "theoretical_loss": 4.120013848270222, "tokens_seen": 316866560 }, { "epoch": 0.1, "learning_rate": 0.00045654964894684054, "loss": 3.1744, "theoretical_loss": 4.119914768867536, "tokens_seen": 316932096 }, { "epoch": 0.1, "learning_rate": 0.00045653961885656967, "loss": 2.9482, "theoretical_loss": 4.11981571568587, "tokens_seen": 316997632 }, { "epoch": 0.1, "learning_rate": 0.0004565295887662989, "loss": 3.1156, "theoretical_loss": 4.119716688712866, "tokens_seen": 317063168 }, { "epoch": 0.1, "learning_rate": 0.0004565195586760281, "loss": 3.0127, "theoretical_loss": 4.119617687936175, "tokens_seen": 317128704 }, { "epoch": 0.1, "learning_rate": 0.00045650952858575727, "loss": 2.9893, "theoretical_loss": 4.119518713343455, "tokens_seen": 317194240 }, { "epoch": 0.1, "learning_rate": 0.00045649949849548645, "loss": 2.9135, "theoretical_loss": 4.119419764922374, "tokens_seen": 317259776 }, { "epoch": 0.1, "learning_rate": 0.0004564894684052157, "loss": 2.9428, "theoretical_loss": 4.119320842660606, "tokens_seen": 317325312 }, { "epoch": 0.1, "learning_rate": 0.0004564794383149448, "loss": 3.0893, "theoretical_loss": 4.119221946545836, "tokens_seen": 317390848 }, { "epoch": 0.1, "learning_rate": 0.00045646940822467405, "loss": 2.954, "theoretical_loss": 4.119123076565755, "tokens_seen": 317456384 }, { "epoch": 0.1, "learning_rate": 0.00045645937813440317, "loss": 3.1031, "theoretical_loss": 4.119024232708064, "tokens_seen": 317521920 }, { "epoch": 0.1, "learning_rate": 0.0004564493480441324, "loss": 3.1224, "theoretical_loss": 4.118925414960472, "tokens_seen": 317587456 }, { "epoch": 0.1, "learning_rate": 0.00045643931795386164, "loss": 3.1016, "theoretical_loss": 4.118826623310696, "tokens_seen": 317652992 }, { "epoch": 0.1, "learning_rate": 0.00045642928786359077, "loss": 3.0232, "theoretical_loss": 4.11872785774646, "tokens_seen": 317718528 }, { "epoch": 0.1, "learning_rate": 0.00045641925777332, "loss": 3.211, "theoretical_loss": 4.1186291182555, "tokens_seen": 317784064 }, { "epoch": 0.1, "objective/train/docs_used": 191330, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.303915023803711, "objective/train/theoretical_loss": 4.118530404825556, "objective/train/tokens_used": 338309600, "theoretical_loss": 4.118530404825556, "tokens_seen": 317849600 }, { "epoch": 0.1, "learning_rate": 0.00045640922768304913, "loss": 3.1621, "theoretical_loss": 4.118530404825556, "tokens_seen": 317849600 }, { "epoch": 0.1, "learning_rate": 0.00045639919759277837, "loss": 3.2408, "theoretical_loss": 4.11843171744438, "tokens_seen": 317915136 }, { "epoch": 0.1, "learning_rate": 0.00045638916750250755, "loss": 3.0122, "theoretical_loss": 4.118333056099728, "tokens_seen": 317980672 }, { "epoch": 0.1, "learning_rate": 0.00045637913741223673, "loss": 3.3026, "theoretical_loss": 4.11823442077937, "tokens_seen": 318046208 }, { "epoch": 0.1, "learning_rate": 0.0004563691073219659, "loss": 3.1502, "theoretical_loss": 4.1181358114710775, "tokens_seen": 318111744 }, { "epoch": 0.1, "learning_rate": 0.00045635907723169515, "loss": 3.1201, "theoretical_loss": 4.1180372281626365, "tokens_seen": 318177280 }, { "epoch": 0.1, "learning_rate": 0.00045634904714142427, "loss": 2.9375, "theoretical_loss": 4.117938670841838, "tokens_seen": 318242816 }, { "epoch": 0.1, "learning_rate": 0.0004563390170511535, "loss": 3.2836, "theoretical_loss": 4.117840139496482, "tokens_seen": 318308352 }, { "epoch": 0.1, "learning_rate": 0.00045632898696088263, "loss": 3.2809, "theoretical_loss": 4.117741634114376, "tokens_seen": 318373888 }, { "epoch": 0.1, "learning_rate": 0.00045631895687061187, "loss": 3.3275, "theoretical_loss": 4.1176431546833365, "tokens_seen": 318439424 }, { "epoch": 0.1, "learning_rate": 0.00045630892678034105, "loss": 3.2046, "theoretical_loss": 4.117544701191187, "tokens_seen": 318504960 }, { "epoch": 0.1, "learning_rate": 0.00045629889669007023, "loss": 2.7935, "theoretical_loss": 4.117446273625763, "tokens_seen": 318570496 }, { "epoch": 0.1, "learning_rate": 0.0004562888665997994, "loss": 3.1972, "theoretical_loss": 4.117347871974903, "tokens_seen": 318636032 }, { "epoch": 0.1, "learning_rate": 0.0004562788365095286, "loss": 3.283, "theoretical_loss": 4.1172494962264565, "tokens_seen": 318701568 }, { "epoch": 0.1, "learning_rate": 0.0004562688064192578, "loss": 3.0075, "theoretical_loss": 4.117151146368282, "tokens_seen": 318767104 }, { "epoch": 0.1, "learning_rate": 0.000456258776328987, "loss": 3.2145, "theoretical_loss": 4.117052822388243, "tokens_seen": 318832640 }, { "epoch": 0.1, "learning_rate": 0.00045624874623871614, "loss": 2.6977, "theoretical_loss": 4.116954524274216, "tokens_seen": 318898176 }, { "epoch": 0.1, "learning_rate": 0.0004562387161484454, "loss": 3.2573, "theoretical_loss": 4.11685625201408, "tokens_seen": 318963712 }, { "epoch": 0.1, "learning_rate": 0.0004562286860581745, "loss": 2.9499, "theoretical_loss": 4.116758005595727, "tokens_seen": 319029248 }, { "epoch": 0.1, "learning_rate": 0.00045621865596790374, "loss": 2.9781, "theoretical_loss": 4.116659785007055, "tokens_seen": 319094784 }, { "epoch": 0.1, "learning_rate": 0.0004562086258776329, "loss": 3.1879, "theoretical_loss": 4.116561590235969, "tokens_seen": 319160320 }, { "epoch": 0.1, "learning_rate": 0.0004561985957873621, "loss": 3.1697, "theoretical_loss": 4.116463421270385, "tokens_seen": 319225856 }, { "epoch": 0.1, "learning_rate": 0.0004561885656970913, "loss": 2.9933, "theoretical_loss": 4.116365278098225, "tokens_seen": 319291392 }, { "epoch": 0.1, "learning_rate": 0.0004561785356068205, "loss": 3.1033, "theoretical_loss": 4.116267160707421, "tokens_seen": 319356928 }, { "epoch": 0.1, "learning_rate": 0.00045616850551654964, "loss": 2.9044, "theoretical_loss": 4.11616906908591, "tokens_seen": 319422464 }, { "epoch": 0.1, "objective/train/docs_used": 192764, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9580471515655518, "objective/train/theoretical_loss": 4.11607100322164, "objective/train/tokens_used": 339948000, "theoretical_loss": 4.11607100322164, "tokens_seen": 319488000 }, { "epoch": 0.1, "learning_rate": 0.0004561584754262789, "loss": 2.9887, "theoretical_loss": 4.11607100322164, "tokens_seen": 319488000 }, { "epoch": 0.1, "learning_rate": 0.000456148445336008, "loss": 3.0573, "theoretical_loss": 4.115972963102565, "tokens_seen": 319553536 }, { "epoch": 0.1, "learning_rate": 0.00045613841524573724, "loss": 3.2024, "theoretical_loss": 4.11587494871665, "tokens_seen": 319619072 }, { "epoch": 0.1, "learning_rate": 0.0004561283851554664, "loss": 3.2376, "theoretical_loss": 4.115776960051864, "tokens_seen": 319684608 }, { "epoch": 0.1, "learning_rate": 0.0004561183550651956, "loss": 3.0396, "theoretical_loss": 4.11567899709619, "tokens_seen": 319750144 }, { "epoch": 0.1, "learning_rate": 0.0004561083249749248, "loss": 2.7815, "theoretical_loss": 4.115581059837612, "tokens_seen": 319815680 }, { "epoch": 0.1, "learning_rate": 0.00045609829488465396, "loss": 3.1572, "theoretical_loss": 4.115483148264127, "tokens_seen": 319881216 }, { "epoch": 0.1, "learning_rate": 0.00045608826479438314, "loss": 3.0449, "theoretical_loss": 4.115385262363739, "tokens_seen": 319946752 }, { "epoch": 0.1, "learning_rate": 0.0004560782347041124, "loss": 3.0447, "theoretical_loss": 4.1152874021244585, "tokens_seen": 320012288 }, { "epoch": 0.1, "learning_rate": 0.0004560682046138415, "loss": 2.905, "theoretical_loss": 4.115189567534307, "tokens_seen": 320077824 }, { "epoch": 0.1, "learning_rate": 0.00045605817452357074, "loss": 3.0628, "theoretical_loss": 4.115091758581309, "tokens_seen": 320143360 }, { "epoch": 0.1, "learning_rate": 0.00045604814443329987, "loss": 2.9603, "theoretical_loss": 4.114993975253505, "tokens_seen": 320208896 }, { "epoch": 0.1, "learning_rate": 0.0004560381143430291, "loss": 2.9837, "theoretical_loss": 4.114896217538935, "tokens_seen": 320274432 }, { "epoch": 0.1, "learning_rate": 0.0004560280842527583, "loss": 2.8237, "theoretical_loss": 4.114798485425652, "tokens_seen": 320339968 }, { "epoch": 0.1, "learning_rate": 0.00045601805416248747, "loss": 3.0052, "theoretical_loss": 4.114700778901717, "tokens_seen": 320405504 }, { "epoch": 0.1, "learning_rate": 0.00045600802407221665, "loss": 3.0487, "theoretical_loss": 4.114603097955197, "tokens_seen": 320471040 }, { "epoch": 0.1, "learning_rate": 0.0004559979939819459, "loss": 3.2513, "theoretical_loss": 4.114505442574167, "tokens_seen": 320536576 }, { "epoch": 0.1, "learning_rate": 0.000455987963891675, "loss": 3.2844, "theoretical_loss": 4.1144078127467125, "tokens_seen": 320602112 }, { "epoch": 0.1, "learning_rate": 0.00045597793380140425, "loss": 2.7771, "theoretical_loss": 4.114310208460924, "tokens_seen": 320667648 }, { "epoch": 0.1, "learning_rate": 0.00045596790371113337, "loss": 3.1613, "theoretical_loss": 4.114212629704902, "tokens_seen": 320733184 }, { "epoch": 0.1, "learning_rate": 0.0004559578736208626, "loss": 2.9396, "theoretical_loss": 4.114115076466755, "tokens_seen": 320798720 }, { "epoch": 0.1, "learning_rate": 0.0004559478435305918, "loss": 2.9128, "theoretical_loss": 4.114017548734598, "tokens_seen": 320864256 }, { "epoch": 0.1, "learning_rate": 0.00045593781344032097, "loss": 2.9857, "theoretical_loss": 4.113920046496554, "tokens_seen": 320929792 }, { "epoch": 0.1, "learning_rate": 0.00045592778335005015, "loss": 3.1028, "theoretical_loss": 4.113822569740757, "tokens_seen": 320995328 }, { "epoch": 0.1, "learning_rate": 0.00045591775325977933, "loss": 3.5144, "theoretical_loss": 4.113725118455344, "tokens_seen": 321060864 }, { "epoch": 0.1, "objective/train/docs_used": 193449, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.532770872116089, "objective/train/theoretical_loss": 4.113627692628464, "objective/train/tokens_used": 341586400, "theoretical_loss": 4.113627692628464, "tokens_seen": 321126400 }, { "epoch": 0.1, "learning_rate": 0.0004559077231695085, "loss": 3.0643, "theoretical_loss": 4.113627692628464, "tokens_seen": 321126400 }, { "epoch": 0.1, "learning_rate": 0.00045589769307923775, "loss": 3.1924, "theoretical_loss": 4.113530292248273, "tokens_seen": 321191936 }, { "epoch": 0.1, "learning_rate": 0.0004558876629889669, "loss": 2.9817, "theoretical_loss": 4.113432917302934, "tokens_seen": 321257472 }, { "epoch": 0.1, "learning_rate": 0.0004558776328986961, "loss": 2.954, "theoretical_loss": 4.113335567780618, "tokens_seen": 321323008 }, { "epoch": 0.1, "learning_rate": 0.00045586760280842524, "loss": 2.9283, "theoretical_loss": 4.113238243669504, "tokens_seen": 321388544 }, { "epoch": 0.1, "learning_rate": 0.0004558575727181545, "loss": 3.0588, "theoretical_loss": 4.113140944957781, "tokens_seen": 321454080 }, { "epoch": 0.1, "learning_rate": 0.00045584754262788365, "loss": 2.9978, "theoretical_loss": 4.113043671633641, "tokens_seen": 321519616 }, { "epoch": 0.1, "learning_rate": 0.00045583751253761284, "loss": 3.1526, "theoretical_loss": 4.11294642368529, "tokens_seen": 321585152 }, { "epoch": 0.1, "learning_rate": 0.000455827482447342, "loss": 2.856, "theoretical_loss": 4.112849201100938, "tokens_seen": 321650688 }, { "epoch": 0.1, "learning_rate": 0.00045581745235707125, "loss": 3.0283, "theoretical_loss": 4.1127520038688035, "tokens_seen": 321716224 }, { "epoch": 0.1, "learning_rate": 0.0004558074222668004, "loss": 3.1706, "theoretical_loss": 4.112654831977112, "tokens_seen": 321781760 }, { "epoch": 0.1, "learning_rate": 0.0004557973921765296, "loss": 3.1208, "theoretical_loss": 4.1125576854141, "tokens_seen": 321847296 }, { "epoch": 0.1, "learning_rate": 0.00045578736208625874, "loss": 3.2345, "theoretical_loss": 4.112460564168009, "tokens_seen": 321912832 }, { "epoch": 0.1, "learning_rate": 0.000455777331995988, "loss": 3.0428, "theoretical_loss": 4.112363468227088, "tokens_seen": 321978368 }, { "epoch": 0.1, "learning_rate": 0.00045576730190571716, "loss": 3.3738, "theoretical_loss": 4.112266397579598, "tokens_seen": 322043904 }, { "epoch": 0.1, "learning_rate": 0.00045575727181544634, "loss": 3.0012, "theoretical_loss": 4.112169352213801, "tokens_seen": 322109440 }, { "epoch": 0.1, "learning_rate": 0.0004557472417251755, "loss": 2.7209, "theoretical_loss": 4.1120723321179735, "tokens_seen": 322174976 }, { "epoch": 0.1, "learning_rate": 0.0004557372116349047, "loss": 3.0685, "theoretical_loss": 4.111975337280397, "tokens_seen": 322240512 }, { "epoch": 0.1, "learning_rate": 0.0004557271815446339, "loss": 3.2726, "theoretical_loss": 4.111878367689359, "tokens_seen": 322306048 }, { "epoch": 0.1, "learning_rate": 0.0004557171514543631, "loss": 2.9605, "theoretical_loss": 4.1117814233331575, "tokens_seen": 322371584 }, { "epoch": 0.1, "learning_rate": 0.00045570712136409224, "loss": 3.0389, "theoretical_loss": 4.111684504200099, "tokens_seen": 322437120 }, { "epoch": 0.1, "learning_rate": 0.0004556970912738215, "loss": 3.0147, "theoretical_loss": 4.111587610278494, "tokens_seen": 322502656 }, { "epoch": 0.1, "learning_rate": 0.0004556870611835507, "loss": 3.126, "theoretical_loss": 4.111490741556663, "tokens_seen": 322568192 }, { "epoch": 0.1, "learning_rate": 0.00045567703109327984, "loss": 3.2523, "theoretical_loss": 4.1113938980229365, "tokens_seen": 322633728 }, { "epoch": 0.1, "learning_rate": 0.0004556670010030091, "loss": 2.9724, "theoretical_loss": 4.11129707966565, "tokens_seen": 322699264 }, { "epoch": 0.1, "objective/train/docs_used": 194727, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2888762950897217, "objective/train/theoretical_loss": 4.111200286473145, "objective/train/tokens_used": 343224800, "theoretical_loss": 4.111200286473145, "tokens_seen": 322764800 }, { "epoch": 0.1, "learning_rate": 0.0004556569709127382, "loss": 3.1618, "theoretical_loss": 4.111200286473145, "tokens_seen": 322764800 }, { "epoch": 0.1, "learning_rate": 0.00045564694082246744, "loss": 2.8571, "theoretical_loss": 4.111103518433776, "tokens_seen": 322830336 }, { "epoch": 0.1, "learning_rate": 0.0004556369107321966, "loss": 3.1171, "theoretical_loss": 4.111006775535901, "tokens_seen": 322895872 }, { "epoch": 0.1, "learning_rate": 0.0004556268806419258, "loss": 3.0964, "theoretical_loss": 4.110910057767887, "tokens_seen": 322961408 }, { "epoch": 0.1, "learning_rate": 0.000455616850551655, "loss": 3.2273, "theoretical_loss": 4.110813365118109, "tokens_seen": 323026944 }, { "epoch": 0.1, "learning_rate": 0.00045560682046138416, "loss": 3.0086, "theoretical_loss": 4.110716697574951, "tokens_seen": 323092480 }, { "epoch": 0.1, "learning_rate": 0.00045559679037111334, "loss": 2.9121, "theoretical_loss": 4.110620055126802, "tokens_seen": 323158016 }, { "epoch": 0.1, "learning_rate": 0.0004555867602808426, "loss": 3.1999, "theoretical_loss": 4.110523437762059, "tokens_seen": 323223552 }, { "epoch": 0.1, "learning_rate": 0.0004555767301905717, "loss": 3.254, "theoretical_loss": 4.11042684546913, "tokens_seen": 323289088 }, { "epoch": 0.1, "learning_rate": 0.00045556670010030094, "loss": 3.0221, "theoretical_loss": 4.110330278236427, "tokens_seen": 323354624 }, { "epoch": 0.1, "learning_rate": 0.00045555667001003007, "loss": 2.9684, "theoretical_loss": 4.110233736052372, "tokens_seen": 323420160 }, { "epoch": 0.1, "learning_rate": 0.0004555466399197593, "loss": 3.2605, "theoretical_loss": 4.110137218905393, "tokens_seen": 323485696 }, { "epoch": 0.1, "learning_rate": 0.0004555366098294885, "loss": 3.1783, "theoretical_loss": 4.110040726783927, "tokens_seen": 323551232 }, { "epoch": 0.1, "learning_rate": 0.00045552657973921767, "loss": 3.0009, "theoretical_loss": 4.109944259676419, "tokens_seen": 323616768 }, { "epoch": 0.1, "learning_rate": 0.00045551654964894685, "loss": 2.9394, "theoretical_loss": 4.109847817571319, "tokens_seen": 323682304 }, { "epoch": 0.1, "learning_rate": 0.0004555065195586761, "loss": 3.0756, "theoretical_loss": 4.109751400457089, "tokens_seen": 323747840 }, { "epoch": 0.1, "learning_rate": 0.0004554964894684052, "loss": 3.1542, "theoretical_loss": 4.109655008322195, "tokens_seen": 323813376 }, { "epoch": 0.1, "learning_rate": 0.00045548645937813445, "loss": 3.0225, "theoretical_loss": 4.109558641155112, "tokens_seen": 323878912 }, { "epoch": 0.1, "learning_rate": 0.00045547642928786357, "loss": 3.1003, "theoretical_loss": 4.109462298944322, "tokens_seen": 323944448 }, { "epoch": 0.1, "learning_rate": 0.0004554663991975928, "loss": 3.1359, "theoretical_loss": 4.109365981678316, "tokens_seen": 324009984 }, { "epoch": 0.1, "learning_rate": 0.000455456369107322, "loss": 3.0616, "theoretical_loss": 4.109269689345592, "tokens_seen": 324075520 }, { "epoch": 0.1, "learning_rate": 0.00045544633901705117, "loss": 3.0413, "theoretical_loss": 4.109173421934654, "tokens_seen": 324141056 }, { "epoch": 0.1, "learning_rate": 0.00045543630892678035, "loss": 3.1208, "theoretical_loss": 4.109077179434016, "tokens_seen": 324206592 }, { "epoch": 0.1, "learning_rate": 0.00045542627883650953, "loss": 3.1749, "theoretical_loss": 4.1089809618321995, "tokens_seen": 324272128 }, { "epoch": 0.1, "learning_rate": 0.0004554162487462387, "loss": 2.6465, "theoretical_loss": 4.108884769117731, "tokens_seen": 324337664 }, { "epoch": 0.1, "objective/train/docs_used": 195463, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8933980464935303, "objective/train/theoretical_loss": 4.108788601279149, "objective/train/tokens_used": 344863200, "theoretical_loss": 4.108788601279149, "tokens_seen": 324403200 }, { "epoch": 0.1, "learning_rate": 0.00045540621865596795, "loss": 3.0332, "theoretical_loss": 4.108788601279149, "tokens_seen": 324403200 }, { "epoch": 0.1, "learning_rate": 0.0004553961885656971, "loss": 2.9037, "theoretical_loss": 4.108692458304994, "tokens_seen": 324468736 }, { "epoch": 0.1, "learning_rate": 0.0004553861584754263, "loss": 3.2058, "theoretical_loss": 4.108596340183819, "tokens_seen": 324534272 }, { "epoch": 0.1, "learning_rate": 0.00045537612838515544, "loss": 3.1413, "theoretical_loss": 4.108500246904184, "tokens_seen": 324599808 }, { "epoch": 0.1, "learning_rate": 0.0004553660982948847, "loss": 3.1149, "theoretical_loss": 4.108404178454651, "tokens_seen": 324665344 }, { "epoch": 0.1, "learning_rate": 0.00045535606820461385, "loss": 2.7925, "theoretical_loss": 4.1083081348237975, "tokens_seen": 324730880 }, { "epoch": 0.1, "learning_rate": 0.00045534603811434304, "loss": 3.0427, "theoretical_loss": 4.108212116000203, "tokens_seen": 324796416 }, { "epoch": 0.1, "learning_rate": 0.0004553360080240722, "loss": 2.9514, "theoretical_loss": 4.108116121972457, "tokens_seen": 324861952 }, { "epoch": 0.1, "learning_rate": 0.00045532597793380145, "loss": 3.0382, "theoretical_loss": 4.108020152729157, "tokens_seen": 324927488 }, { "epoch": 0.1, "learning_rate": 0.0004553159478435306, "loss": 3.0567, "theoretical_loss": 4.107924208258905, "tokens_seen": 324993024 }, { "epoch": 0.1, "learning_rate": 0.0004553059177532598, "loss": 3.0865, "theoretical_loss": 4.107828288550314, "tokens_seen": 325058560 }, { "epoch": 0.1, "learning_rate": 0.00045529588766298894, "loss": 3.1368, "theoretical_loss": 4.107732393592003, "tokens_seen": 325124096 }, { "epoch": 0.1, "learning_rate": 0.0004552858575727182, "loss": 2.9862, "theoretical_loss": 4.107636523372598, "tokens_seen": 325189632 }, { "epoch": 0.1, "learning_rate": 0.00045527582748244736, "loss": 3.0282, "theoretical_loss": 4.107540677880734, "tokens_seen": 325255168 }, { "epoch": 0.1, "learning_rate": 0.00045526579739217654, "loss": 3.1162, "theoretical_loss": 4.107444857105052, "tokens_seen": 325320704 }, { "epoch": 0.1, "learning_rate": 0.0004552557673019057, "loss": 3.1163, "theoretical_loss": 4.107349061034201, "tokens_seen": 325386240 }, { "epoch": 0.1, "learning_rate": 0.0004552457372116349, "loss": 2.8527, "theoretical_loss": 4.107253289656838, "tokens_seen": 325451776 }, { "epoch": 0.1, "learning_rate": 0.0004552357071213641, "loss": 3.0606, "theoretical_loss": 4.107157542961628, "tokens_seen": 325517312 }, { "epoch": 0.1, "learning_rate": 0.0004552256770310933, "loss": 3.0527, "theoretical_loss": 4.10706182093724, "tokens_seen": 325582848 }, { "epoch": 0.1, "learning_rate": 0.00045521564694082244, "loss": 3.4321, "theoretical_loss": 4.106966123572356, "tokens_seen": 325648384 }, { "epoch": 0.1, "learning_rate": 0.0004552056168505517, "loss": 2.9409, "theoretical_loss": 4.106870450855661, "tokens_seen": 325713920 }, { "epoch": 0.1, "learning_rate": 0.0004551955867602808, "loss": 3.1752, "theoretical_loss": 4.106774802775849, "tokens_seen": 325779456 }, { "epoch": 0.1, "learning_rate": 0.00045518555667001004, "loss": 3.0561, "theoretical_loss": 4.106679179321622, "tokens_seen": 325844992 }, { "epoch": 0.1, "learning_rate": 0.0004551755265797392, "loss": 3.091, "theoretical_loss": 4.106583580481689, "tokens_seen": 325910528 }, { "epoch": 0.1, "learning_rate": 0.0004551654964894684, "loss": 3.0422, "theoretical_loss": 4.106488006244767, "tokens_seen": 325976064 }, { "epoch": 0.1, "objective/train/docs_used": 196731, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.885859727859497, "objective/train/theoretical_loss": 4.106392456599577, "objective/train/tokens_used": 346501600, "theoretical_loss": 4.106392456599577, "tokens_seen": 326041600 }, { "epoch": 0.1, "learning_rate": 0.0004551554663991976, "loss": 2.9643, "theoretical_loss": 4.106392456599577, "tokens_seen": 326041600 }, { "epoch": 0.1, "learning_rate": 0.0004551454363089268, "loss": 3.112, "theoretical_loss": 4.106296931534854, "tokens_seen": 326107136 }, { "epoch": 0.1, "learning_rate": 0.00045513540621865595, "loss": 2.9576, "theoretical_loss": 4.106201431039335, "tokens_seen": 326172672 }, { "epoch": 0.1, "learning_rate": 0.0004551253761283852, "loss": 2.6294, "theoretical_loss": 4.106105955101766, "tokens_seen": 326238208 }, { "epoch": 0.1, "learning_rate": 0.0004551153460381143, "loss": 3.1334, "theoretical_loss": 4.1060105037109, "tokens_seen": 326303744 }, { "epoch": 0.1, "learning_rate": 0.00045510531594784354, "loss": 2.9134, "theoretical_loss": 4.105915076855499, "tokens_seen": 326369280 }, { "epoch": 0.1, "learning_rate": 0.0004550952858575727, "loss": 3.0419, "theoretical_loss": 4.105819674524332, "tokens_seen": 326434816 }, { "epoch": 0.1, "learning_rate": 0.0004550852557673019, "loss": 3.009, "theoretical_loss": 4.105724296706172, "tokens_seen": 326500352 }, { "epoch": 0.1, "learning_rate": 0.0004550752256770311, "loss": 2.9717, "theoretical_loss": 4.105628943389805, "tokens_seen": 326565888 }, { "epoch": 0.1, "learning_rate": 0.00045506519558676027, "loss": 2.9638, "theoretical_loss": 4.1055336145640196, "tokens_seen": 326631424 }, { "epoch": 0.1, "learning_rate": 0.00045505516549648945, "loss": 3.3072, "theoretical_loss": 4.105438310217615, "tokens_seen": 326696960 }, { "epoch": 0.1, "learning_rate": 0.0004550451354062187, "loss": 3.0272, "theoretical_loss": 4.105343030339395, "tokens_seen": 326762496 }, { "epoch": 0.1, "learning_rate": 0.0004550351053159478, "loss": 3.0128, "theoretical_loss": 4.1052477749181735, "tokens_seen": 326828032 }, { "epoch": 0.1, "learning_rate": 0.00045502507522567705, "loss": 3.0124, "theoretical_loss": 4.10515254394277, "tokens_seen": 326893568 }, { "epoch": 0.1, "learning_rate": 0.00045501504513540623, "loss": 2.8748, "theoretical_loss": 4.1050573374020125, "tokens_seen": 326959104 }, { "epoch": 0.1, "learning_rate": 0.0004550050150451354, "loss": 3.0619, "theoretical_loss": 4.104962155284734, "tokens_seen": 327024640 }, { "epoch": 0.1, "learning_rate": 0.0004549949849548646, "loss": 2.8882, "theoretical_loss": 4.104866997579778, "tokens_seen": 327090176 }, { "epoch": 0.1, "learning_rate": 0.00045498495486459377, "loss": 3.0027, "theoretical_loss": 4.104771864275993, "tokens_seen": 327155712 }, { "epoch": 0.1, "learning_rate": 0.00045497492477432295, "loss": 3.1058, "theoretical_loss": 4.104676755362237, "tokens_seen": 327221248 }, { "epoch": 0.1, "learning_rate": 0.0004549648946840522, "loss": 3.0508, "theoretical_loss": 4.104581670827372, "tokens_seen": 327286784 }, { "epoch": 0.1, "learning_rate": 0.00045495486459378137, "loss": 2.8934, "theoretical_loss": 4.10448661066027, "tokens_seen": 327352320 }, { "epoch": 0.1, "learning_rate": 0.00045494483450351055, "loss": 3.1344, "theoretical_loss": 4.104391574849812, "tokens_seen": 327417856 }, { "epoch": 0.1, "learning_rate": 0.00045493480441323973, "loss": 3.0873, "theoretical_loss": 4.10429656338488, "tokens_seen": 327483392 }, { "epoch": 0.1, "learning_rate": 0.0004549247743229689, "loss": 3.0143, "theoretical_loss": 4.104201576254369, "tokens_seen": 327548928 }, { "epoch": 0.1, "learning_rate": 0.00045491474423269815, "loss": 2.9043, "theoretical_loss": 4.10410661344718, "tokens_seen": 327614464 }, { "debugging/Self-BLEU-5": 0.3841532625917073, "debugging/distinct-1-grams": 0.7832216589580511, "debugging/distinct-2-grams": 0.9618292669346826, "debugging/entropy-1-grams": 5.637375639920272, "debugging/entropy-2-grams": 6.413549369763027, "debugging/length": 486.6666666666667, "debugging/num_segments": 9, "epoch": 0.1, "objective/train/docs_used": 197513, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.052098035812378, "objective/train/theoretical_loss": 4.10401167495222, "objective/train/tokens_used": 348140000, "theoretical_loss": 4.10401167495222, "tokens_seen": 327680000 }, { "epoch": 0.1, "learning_rate": 0.0004549047141424273, "loss": 3.1254, "theoretical_loss": 4.10401167495222, "tokens_seen": 327680000 }, { "epoch": 0.1, "learning_rate": 0.0004548946840521565, "loss": 3.2898, "theoretical_loss": 4.103916760758405, "tokens_seen": 327745536 }, { "epoch": 0.1, "learning_rate": 0.00045488465396188564, "loss": 3.0752, "theoretical_loss": 4.103821870854656, "tokens_seen": 327811072 }, { "epoch": 0.1, "learning_rate": 0.0004548746238716149, "loss": 3.0744, "theoretical_loss": 4.103727005229903, "tokens_seen": 327876608 }, { "epoch": 0.1, "learning_rate": 0.00045486459378134405, "loss": 3.0116, "theoretical_loss": 4.103632163873083, "tokens_seen": 327942144 }, { "epoch": 0.1, "learning_rate": 0.00045485456369107324, "loss": 3.1184, "theoretical_loss": 4.10353734677314, "tokens_seen": 328007680 }, { "epoch": 0.1, "learning_rate": 0.0004548445336008024, "loss": 3.1614, "theoretical_loss": 4.103442553919026, "tokens_seen": 328073216 }, { "epoch": 0.1, "learning_rate": 0.00045483450351053165, "loss": 2.9932, "theoretical_loss": 4.1033477852996985, "tokens_seen": 328138752 }, { "epoch": 0.1, "learning_rate": 0.0004548244734202608, "loss": 3.1685, "theoretical_loss": 4.103253040904124, "tokens_seen": 328204288 }, { "epoch": 0.1, "learning_rate": 0.00045481444332999, "loss": 2.8453, "theoretical_loss": 4.103158320721276, "tokens_seen": 328269824 }, { "epoch": 0.1, "learning_rate": 0.00045480441323971914, "loss": 3.1451, "theoretical_loss": 4.103063624740133, "tokens_seen": 328335360 }, { "epoch": 0.1, "learning_rate": 0.0004547943831494484, "loss": 3.0218, "theoretical_loss": 4.102968952949684, "tokens_seen": 328400896 }, { "epoch": 0.1, "learning_rate": 0.00045478435305917756, "loss": 2.9797, "theoretical_loss": 4.102874305338923, "tokens_seen": 328466432 }, { "epoch": 0.1, "learning_rate": 0.00045477432296890674, "loss": 3.0406, "theoretical_loss": 4.102779681896852, "tokens_seen": 328531968 }, { "epoch": 0.1, "learning_rate": 0.0004547642928786359, "loss": 2.8546, "theoretical_loss": 4.10268508261248, "tokens_seen": 328597504 }, { "epoch": 0.1, "learning_rate": 0.0004547542627883651, "loss": 3.0949, "theoretical_loss": 4.102590507474824, "tokens_seen": 328663040 }, { "epoch": 0.1, "learning_rate": 0.0004547442326980943, "loss": 3.256, "theoretical_loss": 4.1024959564729055, "tokens_seen": 328728576 }, { "epoch": 0.1, "learning_rate": 0.0004547342026078235, "loss": 3.0471, "theoretical_loss": 4.102401429595758, "tokens_seen": 328794112 }, { "epoch": 0.1, "learning_rate": 0.00045472417251755264, "loss": 2.9156, "theoretical_loss": 4.102306926832417, "tokens_seen": 328859648 }, { "epoch": 0.1, "learning_rate": 0.0004547141424272819, "loss": 3.2931, "theoretical_loss": 4.102212448171928, "tokens_seen": 328925184 }, { "epoch": 0.1, "learning_rate": 0.000454704112337011, "loss": 2.982, "theoretical_loss": 4.1021179936033425, "tokens_seen": 328990720 }, { "epoch": 0.1, "learning_rate": 0.00045469408224674024, "loss": 3.0106, "theoretical_loss": 4.102023563115721, "tokens_seen": 329056256 }, { "epoch": 0.1, "learning_rate": 0.0004546840521564694, "loss": 3.0727, "theoretical_loss": 4.10192915669813, "tokens_seen": 329121792 }, { "epoch": 0.1, "learning_rate": 0.0004546740220661986, "loss": 3.1747, "theoretical_loss": 4.1018347743396415, "tokens_seen": 329187328 }, { "epoch": 0.1, "learning_rate": 0.0004546639919759278, "loss": 3.1083, "theoretical_loss": 4.101740416029338, "tokens_seen": 329252864 }, { "epoch": 0.1, "objective/train/docs_used": 198138, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2101240158081055, "objective/train/theoretical_loss": 4.101646081756305, "objective/train/tokens_used": 349778400, "theoretical_loss": 4.101646081756305, "tokens_seen": 329318400 }, { "epoch": 0.1, "learning_rate": 0.000454653961885657, "loss": 3.0282, "theoretical_loss": 4.101646081756305, "tokens_seen": 329318400 }, { "epoch": 0.1, "learning_rate": 0.00045464393179538615, "loss": 2.993, "theoretical_loss": 4.101551771509641, "tokens_seen": 329383936 }, { "epoch": 0.1, "learning_rate": 0.0004546339017051154, "loss": 3.2896, "theoretical_loss": 4.101457485278444, "tokens_seen": 329449472 }, { "epoch": 0.1, "learning_rate": 0.0004546238716148445, "loss": 3.0261, "theoretical_loss": 4.101363223051826, "tokens_seen": 329515008 }, { "epoch": 0.1, "learning_rate": 0.00045461384152457374, "loss": 3.0682, "theoretical_loss": 4.101268984818901, "tokens_seen": 329580544 }, { "epoch": 0.1, "learning_rate": 0.0004546038114343029, "loss": 3.1339, "theoretical_loss": 4.101174770568795, "tokens_seen": 329646080 }, { "epoch": 0.1, "learning_rate": 0.0004545937813440321, "loss": 3.0888, "theoretical_loss": 4.1010805802906365, "tokens_seen": 329711616 }, { "epoch": 0.1, "learning_rate": 0.0004545837512537613, "loss": 3.0345, "theoretical_loss": 4.100986413973564, "tokens_seen": 329777152 }, { "epoch": 0.1, "learning_rate": 0.00045457372116349047, "loss": 3.1313, "theoretical_loss": 4.100892271606721, "tokens_seen": 329842688 }, { "epoch": 0.1, "learning_rate": 0.00045456369107321965, "loss": 2.8485, "theoretical_loss": 4.1007981531792606, "tokens_seen": 329908224 }, { "epoch": 0.1, "learning_rate": 0.0004545536609829489, "loss": 3.19, "theoretical_loss": 4.100704058680341, "tokens_seen": 329973760 }, { "epoch": 0.1, "learning_rate": 0.000454543630892678, "loss": 3.017, "theoretical_loss": 4.1006099880991265, "tokens_seen": 330039296 }, { "epoch": 0.1, "learning_rate": 0.00045453360080240725, "loss": 2.9604, "theoretical_loss": 4.100515941424792, "tokens_seen": 330104832 }, { "epoch": 0.1, "learning_rate": 0.00045452357071213643, "loss": 2.9596, "theoretical_loss": 4.100421918646517, "tokens_seen": 330170368 }, { "epoch": 0.1, "learning_rate": 0.0004545135406218656, "loss": 3.1628, "theoretical_loss": 4.1003279197534885, "tokens_seen": 330235904 }, { "epoch": 0.1, "learning_rate": 0.0004545035105315948, "loss": 2.9544, "theoretical_loss": 4.100233944734899, "tokens_seen": 330301440 }, { "epoch": 0.1, "learning_rate": 0.00045449348044132397, "loss": 3.1729, "theoretical_loss": 4.100139993579952, "tokens_seen": 330366976 }, { "epoch": 0.1, "learning_rate": 0.00045448345035105315, "loss": 2.7687, "theoretical_loss": 4.100046066277853, "tokens_seen": 330432512 }, { "epoch": 0.1, "learning_rate": 0.0004544734202607824, "loss": 3.2236, "theoretical_loss": 4.09995216281782, "tokens_seen": 330498048 }, { "epoch": 0.1, "learning_rate": 0.0004544633901705115, "loss": 3.1535, "theoretical_loss": 4.0998582831890715, "tokens_seen": 330563584 }, { "epoch": 0.1, "learning_rate": 0.00045445336008024075, "loss": 3.11, "theoretical_loss": 4.0997644273808405, "tokens_seen": 330629120 }, { "epoch": 0.1, "learning_rate": 0.0004544433299899699, "loss": 2.8457, "theoretical_loss": 4.09967059538236, "tokens_seen": 330694656 }, { "epoch": 0.1, "learning_rate": 0.0004544332998996991, "loss": 2.9948, "theoretical_loss": 4.099576787182874, "tokens_seen": 330760192 }, { "epoch": 0.1, "learning_rate": 0.0004544232698094283, "loss": 3.1013, "theoretical_loss": 4.099483002771633, "tokens_seen": 330825728 }, { "epoch": 0.1, "learning_rate": 0.0004544132397191575, "loss": 2.9936, "theoretical_loss": 4.099389242137894, "tokens_seen": 330891264 }, { "epoch": 0.1, "objective/train/docs_used": 199416, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0712082386016846, "objective/train/theoretical_loss": 4.099295505270921, "objective/train/tokens_used": 351416800, "theoretical_loss": 4.099295505270921, "tokens_seen": 330956800 }, { "epoch": 0.1, "learning_rate": 0.00045440320962888666, "loss": 3.0119, "theoretical_loss": 4.099295505270921, "tokens_seen": 330956800 }, { "epoch": 0.1, "learning_rate": 0.00045439317953861584, "loss": 2.9404, "theoretical_loss": 4.099201792159985, "tokens_seen": 331022336 }, { "epoch": 0.1, "learning_rate": 0.000454383149448345, "loss": 2.9773, "theoretical_loss": 4.099108102794363, "tokens_seen": 331087872 }, { "epoch": 0.1, "learning_rate": 0.00045437311935807425, "loss": 3.1837, "theoretical_loss": 4.099014437163342, "tokens_seen": 331153408 }, { "epoch": 0.1, "learning_rate": 0.0004543630892678034, "loss": 2.9231, "theoretical_loss": 4.098920795256213, "tokens_seen": 331218944 }, { "epoch": 0.1, "learning_rate": 0.0004543530591775326, "loss": 2.9996, "theoretical_loss": 4.098827177062273, "tokens_seen": 331284480 }, { "epoch": 0.1, "learning_rate": 0.0004543430290872618, "loss": 2.9186, "theoretical_loss": 4.098733582570831, "tokens_seen": 331350016 }, { "epoch": 0.1, "learning_rate": 0.000454332998996991, "loss": 3.0911, "theoretical_loss": 4.098640011771198, "tokens_seen": 331415552 }, { "epoch": 0.1, "learning_rate": 0.00045432296890672016, "loss": 2.9677, "theoretical_loss": 4.098546464652693, "tokens_seen": 331481088 }, { "epoch": 0.1, "learning_rate": 0.00045431293881644934, "loss": 2.8865, "theoretical_loss": 4.098452941204643, "tokens_seen": 331546624 }, { "epoch": 0.1, "learning_rate": 0.0004543029087261785, "loss": 2.9476, "theoretical_loss": 4.098359441416383, "tokens_seen": 331612160 }, { "epoch": 0.1, "learning_rate": 0.00045429287863590776, "loss": 2.9588, "theoretical_loss": 4.0982659652772515, "tokens_seen": 331677696 }, { "epoch": 0.1, "learning_rate": 0.0004542828485456369, "loss": 3.1995, "theoretical_loss": 4.098172512776597, "tokens_seen": 331743232 }, { "epoch": 0.1, "learning_rate": 0.0004542728184553661, "loss": 2.9885, "theoretical_loss": 4.098079083903773, "tokens_seen": 331808768 }, { "epoch": 0.1, "learning_rate": 0.00045426278836509525, "loss": 3.1498, "theoretical_loss": 4.097985678648142, "tokens_seen": 331874304 }, { "epoch": 0.1, "learning_rate": 0.0004542527582748245, "loss": 2.948, "theoretical_loss": 4.09789229699907, "tokens_seen": 331939840 }, { "epoch": 0.1, "learning_rate": 0.00045424272818455366, "loss": 3.1034, "theoretical_loss": 4.097798938945933, "tokens_seen": 332005376 }, { "epoch": 0.1, "learning_rate": 0.00045423269809428284, "loss": 3.143, "theoretical_loss": 4.097705604478112, "tokens_seen": 332070912 }, { "epoch": 0.1, "learning_rate": 0.000454222668004012, "loss": 2.8918, "theoretical_loss": 4.097612293584998, "tokens_seen": 332136448 }, { "epoch": 0.1, "learning_rate": 0.0004542126379137412, "loss": 2.817, "theoretical_loss": 4.0975190062559825, "tokens_seen": 332201984 }, { "epoch": 0.1, "learning_rate": 0.00045420260782347044, "loss": 2.9891, "theoretical_loss": 4.097425742480472, "tokens_seen": 332267520 }, { "epoch": 0.1, "learning_rate": 0.0004541925777331996, "loss": 2.9818, "theoretical_loss": 4.097332502247873, "tokens_seen": 332333056 }, { "epoch": 0.1, "learning_rate": 0.0004541825476429288, "loss": 3.0395, "theoretical_loss": 4.0972392855476025, "tokens_seen": 332398592 }, { "epoch": 0.1, "learning_rate": 0.000454172517552658, "loss": 3.127, "theoretical_loss": 4.097146092369084, "tokens_seen": 332464128 }, { "epoch": 0.1, "learning_rate": 0.0004541624874623872, "loss": 2.8874, "theoretical_loss": 4.097052922701746, "tokens_seen": 332529664 }, { "epoch": 0.1, "objective/train/docs_used": 200524, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4482038021087646, "objective/train/theoretical_loss": 4.0969597765350265, "objective/train/tokens_used": 353055200, "theoretical_loss": 4.0969597765350265, "tokens_seen": 332595200 }, { "epoch": 0.1, "learning_rate": 0.00045415245737211635, "loss": 2.9907, "theoretical_loss": 4.0969597765350265, "tokens_seen": 332595200 }, { "epoch": 0.1, "learning_rate": 0.0004541424272818456, "loss": 3.0728, "theoretical_loss": 4.096866653858368, "tokens_seen": 332660736 }, { "epoch": 0.1, "learning_rate": 0.0004541323971915747, "loss": 2.9973, "theoretical_loss": 4.09677355466122, "tokens_seen": 332726272 }, { "epoch": 0.1, "learning_rate": 0.00045412236710130395, "loss": 2.846, "theoretical_loss": 4.096680478933042, "tokens_seen": 332791808 }, { "epoch": 0.1, "learning_rate": 0.0004541123370110331, "loss": 2.9862, "theoretical_loss": 4.0965874266632945, "tokens_seen": 332857344 }, { "epoch": 0.1, "learning_rate": 0.0004541023069207623, "loss": 3.0369, "theoretical_loss": 4.09649439784145, "tokens_seen": 332922880 }, { "epoch": 0.1, "learning_rate": 0.0004540922768304915, "loss": 3.1214, "theoretical_loss": 4.096401392456988, "tokens_seen": 332988416 }, { "epoch": 0.1, "learning_rate": 0.00045408224674022067, "loss": 3.1405, "theoretical_loss": 4.09630841049939, "tokens_seen": 333053952 }, { "epoch": 0.1, "learning_rate": 0.00045407221664994985, "loss": 2.9794, "theoretical_loss": 4.096215451958146, "tokens_seen": 333119488 }, { "epoch": 0.1, "learning_rate": 0.0004540621865596791, "loss": 3.2064, "theoretical_loss": 4.096122516822757, "tokens_seen": 333185024 }, { "epoch": 0.1, "learning_rate": 0.0004540521564694082, "loss": 3.0618, "theoretical_loss": 4.096029605082726, "tokens_seen": 333250560 }, { "epoch": 0.1, "learning_rate": 0.00045404212637913745, "loss": 3.0452, "theoretical_loss": 4.095936716727564, "tokens_seen": 333316096 }, { "epoch": 0.1, "learning_rate": 0.00045403209628886663, "loss": 3.0753, "theoretical_loss": 4.095843851746791, "tokens_seen": 333381632 }, { "epoch": 0.1, "learning_rate": 0.0004540220661985958, "loss": 3.0818, "theoretical_loss": 4.095751010129929, "tokens_seen": 333447168 }, { "epoch": 0.1, "learning_rate": 0.000454012036108325, "loss": 3.178, "theoretical_loss": 4.095658191866512, "tokens_seen": 333512704 }, { "epoch": 0.1, "learning_rate": 0.00045400200601805417, "loss": 3.0465, "theoretical_loss": 4.0955653969460775, "tokens_seen": 333578240 }, { "epoch": 0.1, "learning_rate": 0.00045399197592778335, "loss": 2.8333, "theoretical_loss": 4.095472625358171, "tokens_seen": 333643776 }, { "epoch": 0.1, "learning_rate": 0.0004539819458375126, "loss": 3.3654, "theoretical_loss": 4.095379877092343, "tokens_seen": 333709312 }, { "epoch": 0.1, "learning_rate": 0.0004539719157472417, "loss": 2.9324, "theoretical_loss": 4.095287152138154, "tokens_seen": 333774848 }, { "epoch": 0.1, "learning_rate": 0.00045396188565697095, "loss": 3.1113, "theoretical_loss": 4.0951944504851685, "tokens_seen": 333840384 }, { "epoch": 0.1, "learning_rate": 0.0004539518555667001, "loss": 3.1439, "theoretical_loss": 4.095101772122959, "tokens_seen": 333905920 }, { "epoch": 0.1, "learning_rate": 0.0004539418254764293, "loss": 3.2737, "theoretical_loss": 4.095009117041102, "tokens_seen": 333971456 }, { "epoch": 0.1, "learning_rate": 0.0004539317953861585, "loss": 3.0585, "theoretical_loss": 4.094916485229186, "tokens_seen": 334036992 }, { "epoch": 0.1, "learning_rate": 0.0004539217652958877, "loss": 2.8934, "theoretical_loss": 4.094823876676802, "tokens_seen": 334102528 }, { "epoch": 0.1, "learning_rate": 0.00045391173520561686, "loss": 2.922, "theoretical_loss": 4.094731291373548, "tokens_seen": 334168064 }, { "epoch": 0.1, "objective/train/docs_used": 201196, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.268803834915161, "objective/train/theoretical_loss": 4.094638729309031, "objective/train/tokens_used": 354693600, "theoretical_loss": 4.094638729309031, "tokens_seen": 334233600 }, { "epoch": 0.1, "learning_rate": 0.00045390170511534604, "loss": 3.0566, "theoretical_loss": 4.094638729309031, "tokens_seen": 334233600 }, { "epoch": 0.1, "learning_rate": 0.0004538916750250752, "loss": 2.9384, "theoretical_loss": 4.094546190472862, "tokens_seen": 334299136 }, { "epoch": 0.1, "learning_rate": 0.00045388164493480445, "loss": 2.7438, "theoretical_loss": 4.09445367485466, "tokens_seen": 334364672 }, { "epoch": 0.1, "learning_rate": 0.0004538716148445336, "loss": 3.1787, "theoretical_loss": 4.09436118244405, "tokens_seen": 334430208 }, { "epoch": 0.1, "learning_rate": 0.0004538615847542628, "loss": 3.0136, "theoretical_loss": 4.094268713230667, "tokens_seen": 334495744 }, { "epoch": 0.1, "learning_rate": 0.000453851554663992, "loss": 3.0249, "theoretical_loss": 4.094176267204148, "tokens_seen": 334561280 }, { "epoch": 0.1, "learning_rate": 0.0004538415245737212, "loss": 3.3932, "theoretical_loss": 4.094083844354137, "tokens_seen": 334626816 }, { "epoch": 0.1, "learning_rate": 0.00045383149448345036, "loss": 3.2013, "theoretical_loss": 4.093991444670289, "tokens_seen": 334692352 }, { "epoch": 0.1, "learning_rate": 0.00045382146439317954, "loss": 2.982, "theoretical_loss": 4.093899068142262, "tokens_seen": 334757888 }, { "epoch": 0.1, "learning_rate": 0.0004538114343029087, "loss": 3.1016, "theoretical_loss": 4.093806714759721, "tokens_seen": 334823424 }, { "epoch": 0.1, "learning_rate": 0.00045380140421263796, "loss": 3.2014, "theoretical_loss": 4.093714384512337, "tokens_seen": 334888960 }, { "epoch": 0.1, "learning_rate": 0.0004537913741223671, "loss": 3.0236, "theoretical_loss": 4.093622077389791, "tokens_seen": 334954496 }, { "epoch": 0.1, "learning_rate": 0.0004537813440320963, "loss": 2.9487, "theoretical_loss": 4.093529793381768, "tokens_seen": 335020032 }, { "epoch": 0.1, "learning_rate": 0.00045377131394182545, "loss": 3.2927, "theoretical_loss": 4.093437532477958, "tokens_seen": 335085568 }, { "epoch": 0.1, "learning_rate": 0.0004537612838515547, "loss": 3.111, "theoretical_loss": 4.093345294668063, "tokens_seen": 335151104 }, { "epoch": 0.1, "learning_rate": 0.00045375125376128386, "loss": 3.19, "theoretical_loss": 4.0932530799417846, "tokens_seen": 335216640 }, { "epoch": 0.1, "learning_rate": 0.00045374122367101304, "loss": 3.2401, "theoretical_loss": 4.093160888288837, "tokens_seen": 335282176 }, { "epoch": 0.1, "learning_rate": 0.0004537311935807422, "loss": 3.1176, "theoretical_loss": 4.0930687196989375, "tokens_seen": 335347712 }, { "epoch": 0.1, "learning_rate": 0.0004537211634904714, "loss": 3.0444, "theoretical_loss": 4.0929765741618125, "tokens_seen": 335413248 }, { "epoch": 0.1, "learning_rate": 0.0004537111334002006, "loss": 3.046, "theoretical_loss": 4.092884451667191, "tokens_seen": 335478784 }, { "epoch": 0.1, "learning_rate": 0.0004537011033099298, "loss": 3.1603, "theoretical_loss": 4.092792352204814, "tokens_seen": 335544320 }, { "epoch": 0.1, "learning_rate": 0.00045369107321965895, "loss": 3.0195, "theoretical_loss": 4.092700275764424, "tokens_seen": 335609856 }, { "epoch": 0.1, "learning_rate": 0.0004536810431293882, "loss": 3.1597, "theoretical_loss": 4.092608222335774, "tokens_seen": 335675392 }, { "epoch": 0.1, "learning_rate": 0.00045367101303911737, "loss": 2.9772, "theoretical_loss": 4.092516191908621, "tokens_seen": 335740928 }, { "epoch": 0.1, "learning_rate": 0.00045366098294884655, "loss": 2.8325, "theoretical_loss": 4.09242418447273, "tokens_seen": 335806464 }, { "epoch": 0.1, "objective/train/docs_used": 202460, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.944324254989624, "objective/train/theoretical_loss": 4.092332200017871, "objective/train/tokens_used": 356332000, "theoretical_loss": 4.092332200017871, "tokens_seen": 335872000 }, { "epoch": 0.1, "learning_rate": 0.00045365095285857573, "loss": 2.9899, "theoretical_loss": 4.092332200017871, "tokens_seen": 335872000 }, { "epoch": 0.1, "learning_rate": 0.0004536409227683049, "loss": 3.0494, "theoretical_loss": 4.092240238533822, "tokens_seen": 335937536 }, { "epoch": 0.1, "learning_rate": 0.0004536308926780341, "loss": 2.8351, "theoretical_loss": 4.092148300010367, "tokens_seen": 336003072 }, { "epoch": 0.1, "learning_rate": 0.0004536208625877633, "loss": 2.9533, "theoretical_loss": 4.092056384437297, "tokens_seen": 336068608 }, { "epoch": 0.1, "learning_rate": 0.00045361083249749245, "loss": 2.9323, "theoretical_loss": 4.091964491804409, "tokens_seen": 336134144 }, { "epoch": 0.1, "learning_rate": 0.0004536008024072217, "loss": 2.9688, "theoretical_loss": 4.091872622101506, "tokens_seen": 336199680 }, { "epoch": 0.1, "learning_rate": 0.0004535907723169508, "loss": 3.0854, "theoretical_loss": 4.091780775318399, "tokens_seen": 336265216 }, { "epoch": 0.1, "learning_rate": 0.00045358074222668005, "loss": 2.9489, "theoretical_loss": 4.091688951444904, "tokens_seen": 336330752 }, { "epoch": 0.1, "learning_rate": 0.00045357071213640923, "loss": 2.8713, "theoretical_loss": 4.091597150470845, "tokens_seen": 336396288 }, { "epoch": 0.1, "learning_rate": 0.0004535606820461384, "loss": 3.1211, "theoretical_loss": 4.091505372386051, "tokens_seen": 336461824 }, { "epoch": 0.1, "learning_rate": 0.0004535506519558676, "loss": 3.0486, "theoretical_loss": 4.091413617180358, "tokens_seen": 336527360 }, { "epoch": 0.1, "learning_rate": 0.00045354062186559683, "loss": 3.3574, "theoretical_loss": 4.091321884843609, "tokens_seen": 336592896 }, { "epoch": 0.1, "learning_rate": 0.00045353059177532596, "loss": 3.184, "theoretical_loss": 4.091230175365653, "tokens_seen": 336658432 }, { "epoch": 0.1, "learning_rate": 0.0004535205616850552, "loss": 3.0774, "theoretical_loss": 4.0911384887363464, "tokens_seen": 336723968 }, { "epoch": 0.1, "learning_rate": 0.0004535105315947843, "loss": 3.1025, "theoretical_loss": 4.091046824945551, "tokens_seen": 336789504 }, { "epoch": 0.1, "learning_rate": 0.00045350050150451355, "loss": 3.087, "theoretical_loss": 4.0909551839831355, "tokens_seen": 336855040 }, { "epoch": 0.1, "learning_rate": 0.00045349047141424274, "loss": 3.155, "theoretical_loss": 4.090863565838974, "tokens_seen": 336920576 }, { "epoch": 0.1, "learning_rate": 0.0004534804413239719, "loss": 2.8842, "theoretical_loss": 4.090771970502948, "tokens_seen": 336986112 }, { "epoch": 0.1, "learning_rate": 0.0004534704112337011, "loss": 2.9869, "theoretical_loss": 4.090680397964947, "tokens_seen": 337051648 }, { "epoch": 0.1, "learning_rate": 0.0004534603811434303, "loss": 3.1412, "theoretical_loss": 4.090588848214865, "tokens_seen": 337117184 }, { "epoch": 0.1, "learning_rate": 0.0004534503510531595, "loss": 2.9635, "theoretical_loss": 4.0904973212426015, "tokens_seen": 337182720 }, { "epoch": 0.1, "learning_rate": 0.0004534403209628887, "loss": 3.0079, "theoretical_loss": 4.090405817038065, "tokens_seen": 337248256 }, { "epoch": 0.1, "learning_rate": 0.0004534302908726179, "loss": 2.8153, "theoretical_loss": 4.090314335591169, "tokens_seen": 337313792 }, { "epoch": 0.1, "learning_rate": 0.00045342026078234706, "loss": 2.8761, "theoretical_loss": 4.090222876891834, "tokens_seen": 337379328 }, { "epoch": 0.1, "learning_rate": 0.00045341023069207624, "loss": 3.2322, "theoretical_loss": 4.090131440929985, "tokens_seen": 337444864 }, { "epoch": 0.1, "objective/train/docs_used": 203033, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.978555679321289, "objective/train/theoretical_loss": 4.090040027695556, "objective/train/tokens_used": 357970400, "theoretical_loss": 4.090040027695556, "tokens_seen": 337510400 }, { "epoch": 0.1, "learning_rate": 0.0004534002006018054, "loss": 3.1313, "theoretical_loss": 4.090040027695556, "tokens_seen": 337510400 }, { "epoch": 0.1, "learning_rate": 0.00045339017051153465, "loss": 3.1773, "theoretical_loss": 4.0899486371784874, "tokens_seen": 337575936 }, { "epoch": 0.1, "learning_rate": 0.0004533801404212638, "loss": 2.9796, "theoretical_loss": 4.089857269368725, "tokens_seen": 337641472 }, { "epoch": 0.1, "learning_rate": 0.000453370110330993, "loss": 3.089, "theoretical_loss": 4.089765924256218, "tokens_seen": 337707008 }, { "epoch": 0.1, "learning_rate": 0.0004533600802407222, "loss": 2.932, "theoretical_loss": 4.089674601830929, "tokens_seen": 337772544 }, { "epoch": 0.1, "learning_rate": 0.0004533500501504514, "loss": 2.9389, "theoretical_loss": 4.08958330208282, "tokens_seen": 337838080 }, { "epoch": 0.1, "learning_rate": 0.00045334002006018056, "loss": 3.0489, "theoretical_loss": 4.089492025001864, "tokens_seen": 337903616 }, { "epoch": 0.1, "learning_rate": 0.00045332998996990974, "loss": 3.142, "theoretical_loss": 4.089400770578038, "tokens_seen": 337969152 }, { "epoch": 0.1, "learning_rate": 0.0004533199598796389, "loss": 2.7704, "theoretical_loss": 4.089309538801327, "tokens_seen": 338034688 }, { "epoch": 0.1, "learning_rate": 0.00045330992978936816, "loss": 3.0028, "theoretical_loss": 4.08921832966172, "tokens_seen": 338100224 }, { "epoch": 0.1, "learning_rate": 0.0004532998996990973, "loss": 3.2315, "theoretical_loss": 4.0891271431492155, "tokens_seen": 338165760 }, { "epoch": 0.1, "learning_rate": 0.0004532898696088265, "loss": 3.1826, "theoretical_loss": 4.089035979253816, "tokens_seen": 338231296 }, { "epoch": 0.1, "learning_rate": 0.00045327983951855565, "loss": 3.1003, "theoretical_loss": 4.08894483796553, "tokens_seen": 338296832 }, { "epoch": 0.1, "learning_rate": 0.0004532698094282849, "loss": 3.1606, "theoretical_loss": 4.0888537192743755, "tokens_seen": 338362368 }, { "epoch": 0.1, "learning_rate": 0.00045325977933801406, "loss": 2.9502, "theoretical_loss": 4.088762623170373, "tokens_seen": 338427904 }, { "epoch": 0.1, "learning_rate": 0.00045324974924774324, "loss": 3.2166, "theoretical_loss": 4.088671549643553, "tokens_seen": 338493440 }, { "epoch": 0.1, "learning_rate": 0.0004532397191574724, "loss": 2.9103, "theoretical_loss": 4.088580498683948, "tokens_seen": 338558976 }, { "epoch": 0.1, "learning_rate": 0.0004532296890672016, "loss": 3.3325, "theoretical_loss": 4.088489470281601, "tokens_seen": 338624512 }, { "epoch": 0.1, "learning_rate": 0.0004532196589769308, "loss": 2.8166, "theoretical_loss": 4.088398464426559, "tokens_seen": 338690048 }, { "epoch": 0.1, "learning_rate": 0.00045320962888666, "loss": 3.1542, "theoretical_loss": 4.088307481108876, "tokens_seen": 338755584 }, { "epoch": 0.1, "learning_rate": 0.00045319959879638915, "loss": 3.0656, "theoretical_loss": 4.088216520318612, "tokens_seen": 338821120 }, { "epoch": 0.1, "learning_rate": 0.0004531895687061184, "loss": 2.9526, "theoretical_loss": 4.0881255820458335, "tokens_seen": 338886656 }, { "epoch": 0.1, "learning_rate": 0.00045317953861584757, "loss": 3.1781, "theoretical_loss": 4.088034666280614, "tokens_seen": 338952192 }, { "epoch": 0.1, "learning_rate": 0.00045316950852557675, "loss": 2.8558, "theoretical_loss": 4.087943773013032, "tokens_seen": 339017728 }, { "epoch": 0.1, "learning_rate": 0.00045315947843530593, "loss": 3.1344, "theoretical_loss": 4.087852902233173, "tokens_seen": 339083264 }, { "epoch": 0.1, "objective/train/docs_used": 204232, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.195103406906128, "objective/train/theoretical_loss": 4.087762053931129, "objective/train/tokens_used": 359608800, "theoretical_loss": 4.087762053931129, "tokens_seen": 339148800 }, { "epoch": 0.1, "learning_rate": 0.0004531494483450351, "loss": 3.0253, "theoretical_loss": 4.087762053931129, "tokens_seen": 339148800 }, { "epoch": 0.1, "learning_rate": 0.0004531394182547643, "loss": 2.7862, "theoretical_loss": 4.087671228096997, "tokens_seen": 339214336 }, { "epoch": 0.1, "learning_rate": 0.0004531293881644935, "loss": 3.0036, "theoretical_loss": 4.087580424720882, "tokens_seen": 339279872 }, { "epoch": 0.1, "learning_rate": 0.00045311935807422265, "loss": 3.0948, "theoretical_loss": 4.087489643792894, "tokens_seen": 339345408 }, { "epoch": 0.1, "learning_rate": 0.0004531093279839519, "loss": 3.1925, "theoretical_loss": 4.08739888530315, "tokens_seen": 339410944 }, { "epoch": 0.1, "learning_rate": 0.000453099297893681, "loss": 3.2017, "theoretical_loss": 4.087308149241774, "tokens_seen": 339476480 }, { "epoch": 0.1, "learning_rate": 0.00045308926780341025, "loss": 3.164, "theoretical_loss": 4.087217435598894, "tokens_seen": 339542016 }, { "epoch": 0.1, "learning_rate": 0.00045307923771313943, "loss": 3.0525, "theoretical_loss": 4.087126744364646, "tokens_seen": 339607552 }, { "epoch": 0.1, "learning_rate": 0.0004530692076228686, "loss": 2.9962, "theoretical_loss": 4.087036075529172, "tokens_seen": 339673088 }, { "epoch": 0.1, "learning_rate": 0.0004530591775325978, "loss": 3.0656, "theoretical_loss": 4.086945429082618, "tokens_seen": 339738624 }, { "epoch": 0.1, "learning_rate": 0.00045304914744232703, "loss": 3.1211, "theoretical_loss": 4.086854805015141, "tokens_seen": 339804160 }, { "epoch": 0.1, "learning_rate": 0.00045303911735205616, "loss": 3.3196, "theoretical_loss": 4.086764203316902, "tokens_seen": 339869696 }, { "epoch": 0.1, "learning_rate": 0.0004530290872617854, "loss": 2.9489, "theoretical_loss": 4.086673623978064, "tokens_seen": 339935232 }, { "epoch": 0.1, "learning_rate": 0.0004530190571715145, "loss": 3.0491, "theoretical_loss": 4.086583066988802, "tokens_seen": 340000768 }, { "epoch": 0.1, "learning_rate": 0.00045300902708124375, "loss": 2.9703, "theoretical_loss": 4.086492532339296, "tokens_seen": 340066304 }, { "epoch": 0.1, "learning_rate": 0.00045299899699097294, "loss": 2.8628, "theoretical_loss": 4.0864020200197295, "tokens_seen": 340131840 }, { "epoch": 0.1, "learning_rate": 0.0004529889669007021, "loss": 3.0217, "theoretical_loss": 4.086311530020296, "tokens_seen": 340197376 }, { "epoch": 0.1, "learning_rate": 0.0004529789368104313, "loss": 3.2726, "theoretical_loss": 4.086221062331192, "tokens_seen": 340262912 }, { "epoch": 0.1, "learning_rate": 0.0004529689067201605, "loss": 2.7499, "theoretical_loss": 4.086130616942621, "tokens_seen": 340328448 }, { "epoch": 0.1, "learning_rate": 0.00045295887662988966, "loss": 3.0251, "theoretical_loss": 4.086040193844794, "tokens_seen": 340393984 }, { "epoch": 0.1, "learning_rate": 0.0004529488465396189, "loss": 2.8855, "theoretical_loss": 4.085949793027927, "tokens_seen": 340459520 }, { "epoch": 0.1, "learning_rate": 0.000452938816449348, "loss": 3.3201, "theoretical_loss": 4.0858594144822415, "tokens_seen": 340525056 }, { "epoch": 0.1, "learning_rate": 0.00045292878635907726, "loss": 3.067, "theoretical_loss": 4.085769058197968, "tokens_seen": 340590592 }, { "epoch": 0.1, "learning_rate": 0.0004529187562688064, "loss": 3.0007, "theoretical_loss": 4.085678724165341, "tokens_seen": 340656128 }, { "epoch": 0.1, "learning_rate": 0.0004529087261785356, "loss": 2.9862, "theoretical_loss": 4.0855884123746, "tokens_seen": 340721664 }, { "epoch": 0.1, "objective/train/docs_used": 205032, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1145529747009277, "objective/train/theoretical_loss": 4.085498122815992, "objective/train/tokens_used": 361247200, "theoretical_loss": 4.085498122815992, "tokens_seen": 340787200 }, { "epoch": 0.1, "learning_rate": 0.0004528986960882648, "loss": 2.8576, "theoretical_loss": 4.085498122815992, "tokens_seen": 340787200 }, { "epoch": 0.1, "learning_rate": 0.000452888665997994, "loss": 2.925, "theoretical_loss": 4.085407855479772, "tokens_seen": 340852736 }, { "epoch": 0.1, "learning_rate": 0.00045287863590772316, "loss": 3.0175, "theoretical_loss": 4.085317610356199, "tokens_seen": 340918272 }, { "epoch": 0.1, "learning_rate": 0.0004528686058174524, "loss": 3.0725, "theoretical_loss": 4.085227387435538, "tokens_seen": 340983808 }, { "epoch": 0.1, "learning_rate": 0.0004528585757271815, "loss": 3.0565, "theoretical_loss": 4.08513718670806, "tokens_seen": 341049344 }, { "epoch": 0.1, "learning_rate": 0.00045284854563691076, "loss": 2.9061, "theoretical_loss": 4.085047008164044, "tokens_seen": 341114880 }, { "epoch": 0.1, "learning_rate": 0.0004528385155466399, "loss": 3.0068, "theoretical_loss": 4.084956851793773, "tokens_seen": 341180416 }, { "epoch": 0.1, "learning_rate": 0.0004528284854563691, "loss": 2.9558, "theoretical_loss": 4.0848667175875395, "tokens_seen": 341245952 }, { "epoch": 0.1, "learning_rate": 0.0004528184553660983, "loss": 3.1372, "theoretical_loss": 4.0847766055356365, "tokens_seen": 341311488 }, { "epoch": 0.1, "learning_rate": 0.0004528084252758275, "loss": 2.8882, "theoretical_loss": 4.084686515628368, "tokens_seen": 341377024 }, { "epoch": 0.1, "learning_rate": 0.00045279839518555667, "loss": 3.1112, "theoretical_loss": 4.0845964478560415, "tokens_seen": 341442560 }, { "epoch": 0.1, "learning_rate": 0.00045278836509528585, "loss": 3.0737, "theoretical_loss": 4.084506402208972, "tokens_seen": 341508096 }, { "epoch": 0.1, "learning_rate": 0.00045277833500501503, "loss": 3.0444, "theoretical_loss": 4.0844163786774805, "tokens_seen": 341573632 }, { "epoch": 0.1, "learning_rate": 0.00045276830491474426, "loss": 3.1, "theoretical_loss": 4.084326377251894, "tokens_seen": 341639168 }, { "epoch": 0.1, "learning_rate": 0.0004527582748244734, "loss": 2.8045, "theoretical_loss": 4.084236397922544, "tokens_seen": 341704704 }, { "epoch": 0.1, "learning_rate": 0.0004527482447342026, "loss": 2.7967, "theoretical_loss": 4.084146440679769, "tokens_seen": 341770240 }, { "epoch": 0.1, "learning_rate": 0.00045273821464393175, "loss": 3.1078, "theoretical_loss": 4.0840565055139155, "tokens_seen": 341835776 }, { "epoch": 0.1, "learning_rate": 0.000452728184553661, "loss": 3.0273, "theoretical_loss": 4.083966592415333, "tokens_seen": 341901312 }, { "epoch": 0.1, "learning_rate": 0.00045271815446339017, "loss": 3.0421, "theoretical_loss": 4.083876701374381, "tokens_seen": 341966848 }, { "epoch": 0.1, "learning_rate": 0.00045270812437311935, "loss": 3.1222, "theoretical_loss": 4.08378683238142, "tokens_seen": 342032384 }, { "epoch": 0.1, "learning_rate": 0.0004526980942828486, "loss": 3.0408, "theoretical_loss": 4.08369698542682, "tokens_seen": 342097920 }, { "epoch": 0.1, "learning_rate": 0.00045268806419257777, "loss": 3.1872, "theoretical_loss": 4.083607160500957, "tokens_seen": 342163456 }, { "epoch": 0.1, "learning_rate": 0.00045267803410230695, "loss": 3.1052, "theoretical_loss": 4.083517357594212, "tokens_seen": 342228992 }, { "epoch": 0.1, "learning_rate": 0.00045266800401203613, "loss": 2.7251, "theoretical_loss": 4.0834275766969705, "tokens_seen": 342294528 }, { "epoch": 0.1, "learning_rate": 0.0004526579739217653, "loss": 3.1931, "theoretical_loss": 4.083337817799629, "tokens_seen": 342360064 }, { "epoch": 0.1, "objective/train/docs_used": 206109, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.671995162963867, "objective/train/theoretical_loss": 4.083248080892584, "objective/train/tokens_used": 362885600, "theoretical_loss": 4.083248080892584, "tokens_seen": 342425600 }, { "epoch": 0.1, "learning_rate": 0.0004526479438314945, "loss": 2.9995, "theoretical_loss": 4.083248080892584, "tokens_seen": 342425600 }, { "epoch": 0.1, "learning_rate": 0.0004526379137412237, "loss": 2.9213, "theoretical_loss": 4.083158365966243, "tokens_seen": 342491136 }, { "epoch": 0.1, "learning_rate": 0.00045262788365095285, "loss": 2.968, "theoretical_loss": 4.083068673011017, "tokens_seen": 342556672 }, { "epoch": 0.1, "learning_rate": 0.0004526178535606821, "loss": 3.0172, "theoretical_loss": 4.082979002017321, "tokens_seen": 342622208 }, { "epoch": 0.1, "learning_rate": 0.0004526078234704112, "loss": 2.7793, "theoretical_loss": 4.082889352975583, "tokens_seen": 342687744 }, { "epoch": 0.1, "learning_rate": 0.00045259779338014045, "loss": 3.0194, "theoretical_loss": 4.082799725876228, "tokens_seen": 342753280 }, { "epoch": 0.1, "learning_rate": 0.00045258776328986963, "loss": 2.8469, "theoretical_loss": 4.082710120709694, "tokens_seen": 342818816 }, { "epoch": 0.1, "learning_rate": 0.0004525777331995988, "loss": 3.0244, "theoretical_loss": 4.082620537466422, "tokens_seen": 342884352 }, { "epoch": 0.1, "learning_rate": 0.000452567703109328, "loss": 3.0037, "theoretical_loss": 4.08253097613686, "tokens_seen": 342949888 }, { "epoch": 0.1, "learning_rate": 0.00045255767301905723, "loss": 3.246, "theoretical_loss": 4.082441436711459, "tokens_seen": 343015424 }, { "epoch": 0.1, "learning_rate": 0.00045254764292878636, "loss": 3.0855, "theoretical_loss": 4.082351919180682, "tokens_seen": 343080960 }, { "epoch": 0.1, "learning_rate": 0.0004525376128385156, "loss": 3.1057, "theoretical_loss": 4.082262423534992, "tokens_seen": 343146496 }, { "epoch": 0.1, "learning_rate": 0.0004525275827482447, "loss": 2.8346, "theoretical_loss": 4.0821729497648604, "tokens_seen": 343212032 }, { "epoch": 0.1, "learning_rate": 0.00045251755265797395, "loss": 2.8814, "theoretical_loss": 4.082083497860766, "tokens_seen": 343277568 }, { "epoch": 0.1, "learning_rate": 0.00045250752256770314, "loss": 2.9733, "theoretical_loss": 4.0819940678131905, "tokens_seen": 343343104 }, { "epoch": 0.1, "learning_rate": 0.0004524974924774323, "loss": 2.9041, "theoretical_loss": 4.081904659612624, "tokens_seen": 343408640 }, { "epoch": 0.1, "learning_rate": 0.0004524874623871615, "loss": 2.9287, "theoretical_loss": 4.081815273249561, "tokens_seen": 343474176 }, { "epoch": 0.1, "learning_rate": 0.0004524774322968907, "loss": 3.0501, "theoretical_loss": 4.081725908714505, "tokens_seen": 343539712 }, { "epoch": 0.1, "learning_rate": 0.00045246740220661986, "loss": 3.0915, "theoretical_loss": 4.081636565997959, "tokens_seen": 343605248 }, { "epoch": 0.1, "learning_rate": 0.0004524573721163491, "loss": 3.2362, "theoretical_loss": 4.08154724509044, "tokens_seen": 343670784 }, { "epoch": 0.1, "learning_rate": 0.0004524473420260782, "loss": 2.734, "theoretical_loss": 4.081457945982465, "tokens_seen": 343736320 }, { "epoch": 0.1, "learning_rate": 0.00045243731193580746, "loss": 2.9276, "theoretical_loss": 4.08136866866456, "tokens_seen": 343801856 }, { "epoch": 0.1, "learning_rate": 0.0004524272818455366, "loss": 3.1926, "theoretical_loss": 4.081279413127254, "tokens_seen": 343867392 }, { "epoch": 0.1, "learning_rate": 0.0004524172517552658, "loss": 3.1345, "theoretical_loss": 4.0811901793610845, "tokens_seen": 343932928 }, { "epoch": 0.1, "learning_rate": 0.000452407221664995, "loss": 3.2088, "theoretical_loss": 4.081100967356594, "tokens_seen": 343998464 }, { "epoch": 0.1, "objective/train/docs_used": 206760, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.804455041885376, "objective/train/theoretical_loss": 4.081011777104333, "objective/train/tokens_used": 364524000, "theoretical_loss": 4.081011777104333, "tokens_seen": 344064000 }, { "epoch": 0.1, "learning_rate": 0.0004523971915747242, "loss": 3.0332, "theoretical_loss": 4.081011777104333, "tokens_seen": 344064000 }, { "epoch": 0.1, "learning_rate": 0.00045238716148445336, "loss": 3.0908, "theoretical_loss": 4.080922608594854, "tokens_seen": 344129536 }, { "epoch": 0.1, "learning_rate": 0.0004523771313941826, "loss": 3.0488, "theoretical_loss": 4.080833461818718, "tokens_seen": 344195072 }, { "epoch": 0.1, "learning_rate": 0.0004523671013039117, "loss": 3.1212, "theoretical_loss": 4.08074433676649, "tokens_seen": 344260608 }, { "epoch": 0.1, "learning_rate": 0.00045235707121364096, "loss": 3.2725, "theoretical_loss": 4.080655233428745, "tokens_seen": 344326144 }, { "epoch": 0.1, "learning_rate": 0.0004523470411233701, "loss": 3.1866, "theoretical_loss": 4.0805661517960585, "tokens_seen": 344391680 }, { "epoch": 0.1, "learning_rate": 0.0004523370110330993, "loss": 2.8178, "theoretical_loss": 4.080477091859015, "tokens_seen": 344457216 }, { "epoch": 0.1, "learning_rate": 0.0004523269809428285, "loss": 3.1468, "theoretical_loss": 4.080388053608205, "tokens_seen": 344522752 }, { "epoch": 0.1, "learning_rate": 0.0004523169508525577, "loss": 2.6369, "theoretical_loss": 4.080299037034224, "tokens_seen": 344588288 }, { "epoch": 0.1, "learning_rate": 0.00045230692076228687, "loss": 3.295, "theoretical_loss": 4.080210042127672, "tokens_seen": 344653824 }, { "epoch": 0.1, "learning_rate": 0.00045229689067201605, "loss": 2.9329, "theoretical_loss": 4.0801210688791585, "tokens_seen": 344719360 }, { "epoch": 0.1, "learning_rate": 0.00045228686058174523, "loss": 2.8028, "theoretical_loss": 4.080032117279296, "tokens_seen": 344784896 }, { "epoch": 0.1, "learning_rate": 0.00045227683049147446, "loss": 3.0295, "theoretical_loss": 4.079943187318703, "tokens_seen": 344850432 }, { "epoch": 0.1, "learning_rate": 0.0004522668004012036, "loss": 2.8921, "theoretical_loss": 4.079854278988005, "tokens_seen": 344915968 }, { "epoch": 0.1, "learning_rate": 0.0004522567703109328, "loss": 2.8358, "theoretical_loss": 4.0797653922778325, "tokens_seen": 344981504 }, { "epoch": 0.1, "learning_rate": 0.00045224674022066195, "loss": 3.1175, "theoretical_loss": 4.0796765271788225, "tokens_seen": 345047040 }, { "epoch": 0.1, "learning_rate": 0.0004522367101303912, "loss": 3.1971, "theoretical_loss": 4.079587683681617, "tokens_seen": 345112576 }, { "epoch": 0.1, "learning_rate": 0.00045222668004012037, "loss": 3.2261, "theoretical_loss": 4.079498861776865, "tokens_seen": 345178112 }, { "epoch": 0.1, "learning_rate": 0.00045221664994984955, "loss": 3.0631, "theoretical_loss": 4.07941006145522, "tokens_seen": 345243648 }, { "epoch": 0.1, "learning_rate": 0.00045220661985957873, "loss": 3.0806, "theoretical_loss": 4.079321282707342, "tokens_seen": 345309184 }, { "epoch": 0.1, "learning_rate": 0.00045219658976930797, "loss": 2.8805, "theoretical_loss": 4.0792325255238975, "tokens_seen": 345374720 }, { "epoch": 0.1, "learning_rate": 0.0004521865596790371, "loss": 3.0152, "theoretical_loss": 4.079143789895556, "tokens_seen": 345440256 }, { "epoch": 0.1, "learning_rate": 0.00045217652958876633, "loss": 2.8819, "theoretical_loss": 4.079055075812997, "tokens_seen": 345505792 }, { "epoch": 0.1, "learning_rate": 0.00045216649949849546, "loss": 3.0413, "theoretical_loss": 4.078966383266904, "tokens_seen": 345571328 }, { "epoch": 0.1, "learning_rate": 0.0004521564694082247, "loss": 2.9147, "theoretical_loss": 4.078877712247963, "tokens_seen": 345636864 }, { "epoch": 0.1, "objective/train/docs_used": 207336, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.845684051513672, "objective/train/theoretical_loss": 4.078789062746872, "objective/train/tokens_used": 366162400, "theoretical_loss": 4.078789062746872, "tokens_seen": 345702400 }, { "epoch": 0.1, "learning_rate": 0.00045214643931795387, "loss": 3.2098, "theoretical_loss": 4.078789062746872, "tokens_seen": 345702400 }, { "epoch": 0.1, "learning_rate": 0.00045213640922768305, "loss": 3.0034, "theoretical_loss": 4.07870043475433, "tokens_seen": 345767936 }, { "epoch": 0.1, "learning_rate": 0.00045212637913741223, "loss": 2.8799, "theoretical_loss": 4.078611828261042, "tokens_seen": 345833472 }, { "epoch": 0.1, "learning_rate": 0.0004521163490471414, "loss": 3.0001, "theoretical_loss": 4.078523243257723, "tokens_seen": 345899008 }, { "epoch": 0.1, "learning_rate": 0.0004521063189568706, "loss": 2.9954, "theoretical_loss": 4.078434679735088, "tokens_seen": 345964544 }, { "epoch": 0.1, "learning_rate": 0.00045209628886659983, "loss": 2.9793, "theoretical_loss": 4.078346137683862, "tokens_seen": 346030080 }, { "epoch": 0.1, "learning_rate": 0.00045208625877632896, "loss": 2.8979, "theoretical_loss": 4.078257617094776, "tokens_seen": 346095616 }, { "epoch": 0.1, "learning_rate": 0.0004520762286860582, "loss": 2.9292, "theoretical_loss": 4.078169117958562, "tokens_seen": 346161152 }, { "epoch": 0.1, "learning_rate": 0.0004520661985957873, "loss": 2.9788, "theoretical_loss": 4.078080640265963, "tokens_seen": 346226688 }, { "epoch": 0.1, "learning_rate": 0.00045205616850551656, "loss": 2.9652, "theoretical_loss": 4.077992184007725, "tokens_seen": 346292224 }, { "epoch": 0.1, "learning_rate": 0.00045204613841524574, "loss": 3.0513, "theoretical_loss": 4.0779037491746, "tokens_seen": 346357760 }, { "epoch": 0.1, "learning_rate": 0.0004520361083249749, "loss": 3.1309, "theoretical_loss": 4.077815335757347, "tokens_seen": 346423296 }, { "epoch": 0.1, "learning_rate": 0.0004520260782347041, "loss": 2.9017, "theoretical_loss": 4.077726943746729, "tokens_seen": 346488832 }, { "epoch": 0.11, "learning_rate": 0.00045201604814443334, "loss": 2.973, "theoretical_loss": 4.077638573133515, "tokens_seen": 346554368 }, { "epoch": 0.11, "learning_rate": 0.00045200601805416246, "loss": 3.3341, "theoretical_loss": 4.077550223908482, "tokens_seen": 346619904 }, { "epoch": 0.11, "learning_rate": 0.0004519959879638917, "loss": 3.0235, "theoretical_loss": 4.0774618960624105, "tokens_seen": 346685440 }, { "epoch": 0.11, "learning_rate": 0.0004519859578736208, "loss": 3.1139, "theoretical_loss": 4.077373589586087, "tokens_seen": 346750976 }, { "epoch": 0.11, "learning_rate": 0.00045197592778335006, "loss": 3.1671, "theoretical_loss": 4.077285304470304, "tokens_seen": 346816512 }, { "epoch": 0.11, "learning_rate": 0.00045196589769307924, "loss": 2.8109, "theoretical_loss": 4.077197040705859, "tokens_seen": 346882048 }, { "epoch": 0.11, "learning_rate": 0.0004519558676028084, "loss": 2.9381, "theoretical_loss": 4.077108798283557, "tokens_seen": 346947584 }, { "epoch": 0.11, "learning_rate": 0.00045194583751253766, "loss": 3.0504, "theoretical_loss": 4.077020577194207, "tokens_seen": 347013120 }, { "epoch": 0.11, "learning_rate": 0.0004519358074222668, "loss": 3.0095, "theoretical_loss": 4.076932377428625, "tokens_seen": 347078656 }, { "epoch": 0.11, "learning_rate": 0.000451925777331996, "loss": 2.9958, "theoretical_loss": 4.076844198977629, "tokens_seen": 347144192 }, { "epoch": 0.11, "learning_rate": 0.0004519157472417252, "loss": 2.8433, "theoretical_loss": 4.076756041832049, "tokens_seen": 347209728 }, { "epoch": 0.11, "learning_rate": 0.0004519057171514544, "loss": 2.9906, "theoretical_loss": 4.076667905982717, "tokens_seen": 347275264 }, { "epoch": 0.11, "objective/train/docs_used": 208796, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0841803550720215, "objective/train/theoretical_loss": 4.076579791420469, "objective/train/tokens_used": 367800800, "theoretical_loss": 4.076579791420469, "tokens_seen": 347340800 }, { "epoch": 0.11, "learning_rate": 0.00045189568706118356, "loss": 3.057, "theoretical_loss": 4.076579791420469, "tokens_seen": 347340800 }, { "epoch": 0.11, "learning_rate": 0.0004518856569709128, "loss": 2.9187, "theoretical_loss": 4.0764916981361505, "tokens_seen": 347406336 }, { "epoch": 0.11, "learning_rate": 0.0004518756268806419, "loss": 2.898, "theoretical_loss": 4.07640362612061, "tokens_seen": 347471872 }, { "epoch": 0.11, "learning_rate": 0.00045186559679037116, "loss": 2.9909, "theoretical_loss": 4.076315575364702, "tokens_seen": 347537408 }, { "epoch": 0.11, "learning_rate": 0.0004518555667001003, "loss": 2.9392, "theoretical_loss": 4.076227545859288, "tokens_seen": 347602944 }, { "epoch": 0.11, "learning_rate": 0.0004518455366098295, "loss": 2.988, "theoretical_loss": 4.076139537595234, "tokens_seen": 347668480 }, { "epoch": 0.11, "learning_rate": 0.0004518355065195587, "loss": 3.1748, "theoretical_loss": 4.076051550563411, "tokens_seen": 347734016 }, { "epoch": 0.11, "learning_rate": 0.0004518254764292879, "loss": 3.1815, "theoretical_loss": 4.075963584754698, "tokens_seen": 347799552 }, { "epoch": 0.11, "learning_rate": 0.00045181544633901707, "loss": 2.9956, "theoretical_loss": 4.075875640159978, "tokens_seen": 347865088 }, { "epoch": 0.11, "learning_rate": 0.00045180541624874625, "loss": 2.8136, "theoretical_loss": 4.075787716770139, "tokens_seen": 347930624 }, { "epoch": 0.11, "learning_rate": 0.00045179538615847543, "loss": 3.1089, "theoretical_loss": 4.075699814576076, "tokens_seen": 347996160 }, { "epoch": 0.11, "learning_rate": 0.00045178535606820466, "loss": 2.8611, "theoretical_loss": 4.075611933568689, "tokens_seen": 348061696 }, { "epoch": 0.11, "learning_rate": 0.0004517753259779338, "loss": 3.0451, "theoretical_loss": 4.075524073738883, "tokens_seen": 348127232 }, { "epoch": 0.11, "learning_rate": 0.000451765295887663, "loss": 2.7604, "theoretical_loss": 4.0754362350775715, "tokens_seen": 348192768 }, { "epoch": 0.11, "learning_rate": 0.00045175526579739215, "loss": 3.1277, "theoretical_loss": 4.075348417575669, "tokens_seen": 348258304 }, { "epoch": 0.11, "learning_rate": 0.0004517452357071214, "loss": 3.0278, "theoretical_loss": 4.075260621224099, "tokens_seen": 348323840 }, { "epoch": 0.11, "learning_rate": 0.00045173520561685057, "loss": 2.9556, "theoretical_loss": 4.075172846013789, "tokens_seen": 348389376 }, { "epoch": 0.11, "learning_rate": 0.00045172517552657975, "loss": 2.7685, "theoretical_loss": 4.075085091935675, "tokens_seen": 348454912 }, { "epoch": 0.11, "learning_rate": 0.00045171514543630893, "loss": 3.0181, "theoretical_loss": 4.074997358980694, "tokens_seen": 348520448 }, { "epoch": 0.11, "learning_rate": 0.00045170511534603817, "loss": 2.9357, "theoretical_loss": 4.0749096471397905, "tokens_seen": 348585984 }, { "epoch": 0.11, "learning_rate": 0.0004516950852557673, "loss": 3.1258, "theoretical_loss": 4.074821956403917, "tokens_seen": 348651520 }, { "epoch": 0.11, "learning_rate": 0.00045168505516549653, "loss": 3.1191, "theoretical_loss": 4.074734286764029, "tokens_seen": 348717056 }, { "epoch": 0.11, "learning_rate": 0.00045167502507522566, "loss": 3.2274, "theoretical_loss": 4.074646638211087, "tokens_seen": 348782592 }, { "epoch": 0.11, "learning_rate": 0.0004516649949849549, "loss": 3.0471, "theoretical_loss": 4.074559010736059, "tokens_seen": 348848128 }, { "epoch": 0.11, "learning_rate": 0.00045165496489468407, "loss": 2.7022, "theoretical_loss": 4.074471404329918, "tokens_seen": 348913664 }, { "epoch": 0.11, "objective/train/docs_used": 209597, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.47709059715271, "objective/train/theoretical_loss": 4.074383818983643, "objective/train/tokens_used": 369439200, "theoretical_loss": 4.074383818983643, "tokens_seen": 348979200 }, { "epoch": 0.11, "learning_rate": 0.00045164493480441325, "loss": 3.1346, "theoretical_loss": 4.074383818983643, "tokens_seen": 348979200 }, { "epoch": 0.11, "learning_rate": 0.00045163490471414244, "loss": 2.8909, "theoretical_loss": 4.074296254688216, "tokens_seen": 349044736 }, { "epoch": 0.11, "learning_rate": 0.0004516248746238716, "loss": 2.9947, "theoretical_loss": 4.074208711434628, "tokens_seen": 349110272 }, { "epoch": 0.11, "learning_rate": 0.0004516148445336008, "loss": 3.012, "theoretical_loss": 4.074121189213873, "tokens_seen": 349175808 }, { "epoch": 0.11, "learning_rate": 0.00045160481444333003, "loss": 3.2239, "theoretical_loss": 4.074033688016953, "tokens_seen": 349241344 }, { "epoch": 0.11, "learning_rate": 0.00045159478435305916, "loss": 3.0182, "theoretical_loss": 4.073946207834872, "tokens_seen": 349306880 }, { "epoch": 0.11, "learning_rate": 0.0004515847542627884, "loss": 2.9586, "theoretical_loss": 4.073858748658644, "tokens_seen": 349372416 }, { "epoch": 0.11, "learning_rate": 0.0004515747241725175, "loss": 3.0998, "theoretical_loss": 4.073771310479284, "tokens_seen": 349437952 }, { "epoch": 0.11, "learning_rate": 0.00045156469408224676, "loss": 3.0868, "theoretical_loss": 4.073683893287816, "tokens_seen": 349503488 }, { "epoch": 0.11, "learning_rate": 0.00045155466399197594, "loss": 2.9732, "theoretical_loss": 4.073596497075267, "tokens_seen": 349569024 }, { "epoch": 0.11, "learning_rate": 0.0004515446339017051, "loss": 2.7333, "theoretical_loss": 4.073509121832672, "tokens_seen": 349634560 }, { "epoch": 0.11, "learning_rate": 0.0004515346038114343, "loss": 2.9115, "theoretical_loss": 4.07342176755107, "tokens_seen": 349700096 }, { "epoch": 0.11, "learning_rate": 0.00045152457372116354, "loss": 3.1394, "theoretical_loss": 4.073334434221506, "tokens_seen": 349765632 }, { "epoch": 0.11, "learning_rate": 0.00045151454363089266, "loss": 3.1035, "theoretical_loss": 4.073247121835028, "tokens_seen": 349831168 }, { "epoch": 0.11, "learning_rate": 0.0004515045135406219, "loss": 3.0943, "theoretical_loss": 4.073159830382695, "tokens_seen": 349896704 }, { "epoch": 0.11, "learning_rate": 0.000451494483450351, "loss": 3.0074, "theoretical_loss": 4.073072559855565, "tokens_seen": 349962240 }, { "epoch": 0.11, "learning_rate": 0.00045148445336008026, "loss": 2.8102, "theoretical_loss": 4.072985310244707, "tokens_seen": 350027776 }, { "epoch": 0.11, "learning_rate": 0.00045147442326980944, "loss": 3.119, "theoretical_loss": 4.072898081541193, "tokens_seen": 350093312 }, { "epoch": 0.11, "learning_rate": 0.0004514643931795386, "loss": 3.0616, "theoretical_loss": 4.0728108737361, "tokens_seen": 350158848 }, { "epoch": 0.11, "learning_rate": 0.0004514543630892678, "loss": 2.9345, "theoretical_loss": 4.0727236868205114, "tokens_seen": 350224384 }, { "epoch": 0.11, "learning_rate": 0.000451444332998997, "loss": 2.7941, "theoretical_loss": 4.072636520785515, "tokens_seen": 350289920 }, { "epoch": 0.11, "learning_rate": 0.00045143430290872617, "loss": 3.1957, "theoretical_loss": 4.072549375622207, "tokens_seen": 350355456 }, { "epoch": 0.11, "learning_rate": 0.0004514242728184554, "loss": 2.6432, "theoretical_loss": 4.072462251321684, "tokens_seen": 350420992 }, { "epoch": 0.11, "learning_rate": 0.00045141424272818453, "loss": 3.1534, "theoretical_loss": 4.072375147875054, "tokens_seen": 350486528 }, { "epoch": 0.11, "learning_rate": 0.00045140421263791376, "loss": 2.9646, "theoretical_loss": 4.072288065273426, "tokens_seen": 350552064 }, { "epoch": 0.11, "objective/train/docs_used": 210243, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.462125539779663, "objective/train/theoretical_loss": 4.0722010035079155, "objective/train/tokens_used": 371077600, "theoretical_loss": 4.0722010035079155, "tokens_seen": 350617600 }, { "epoch": 0.11, "learning_rate": 0.00045139418254764294, "loss": 2.7561, "theoretical_loss": 4.0722010035079155, "tokens_seen": 350617600 }, { "epoch": 0.11, "learning_rate": 0.0004513841524573721, "loss": 2.7177, "theoretical_loss": 4.072113962569645, "tokens_seen": 350683136 }, { "epoch": 0.11, "learning_rate": 0.0004513741223671013, "loss": 2.6755, "theoretical_loss": 4.07202694244974, "tokens_seen": 350748672 }, { "epoch": 0.11, "learning_rate": 0.0004513640922768305, "loss": 2.8439, "theoretical_loss": 4.071939943139334, "tokens_seen": 350814208 }, { "epoch": 0.11, "learning_rate": 0.00045135406218655967, "loss": 2.9904, "theoretical_loss": 4.0718529646295645, "tokens_seen": 350879744 }, { "epoch": 0.11, "learning_rate": 0.0004513440320962889, "loss": 3.0021, "theoretical_loss": 4.071766006911574, "tokens_seen": 350945280 }, { "epoch": 0.11, "learning_rate": 0.00045133400200601803, "loss": 2.9347, "theoretical_loss": 4.071679069976511, "tokens_seen": 351010816 }, { "epoch": 0.11, "learning_rate": 0.00045132397191574727, "loss": 2.9926, "theoretical_loss": 4.07159215381553, "tokens_seen": 351076352 }, { "epoch": 0.11, "learning_rate": 0.0004513139418254764, "loss": 2.9278, "theoretical_loss": 4.07150525841979, "tokens_seen": 351141888 }, { "epoch": 0.11, "learning_rate": 0.00045130391173520563, "loss": 2.9006, "theoretical_loss": 4.071418383780456, "tokens_seen": 351207424 }, { "epoch": 0.11, "learning_rate": 0.0004512938816449348, "loss": 2.9848, "theoretical_loss": 4.071331529888698, "tokens_seen": 351272960 }, { "epoch": 0.11, "learning_rate": 0.000451283851554664, "loss": 2.7971, "theoretical_loss": 4.071244696735692, "tokens_seen": 351338496 }, { "epoch": 0.11, "learning_rate": 0.00045127382146439317, "loss": 2.8099, "theoretical_loss": 4.071157884312619, "tokens_seen": 351404032 }, { "epoch": 0.11, "learning_rate": 0.00045126379137412235, "loss": 3.0982, "theoretical_loss": 4.071071092610664, "tokens_seen": 351469568 }, { "epoch": 0.11, "learning_rate": 0.00045125376128385153, "loss": 3.0576, "theoretical_loss": 4.070984321621021, "tokens_seen": 351535104 }, { "epoch": 0.11, "learning_rate": 0.00045124373119358077, "loss": 2.9455, "theoretical_loss": 4.070897571334886, "tokens_seen": 351600640 }, { "epoch": 0.11, "learning_rate": 0.0004512337011033099, "loss": 3.0845, "theoretical_loss": 4.0708108417434605, "tokens_seen": 351666176 }, { "epoch": 0.11, "learning_rate": 0.00045122367101303913, "loss": 3.0453, "theoretical_loss": 4.070724132837954, "tokens_seen": 351731712 }, { "epoch": 0.11, "learning_rate": 0.0004512136409227683, "loss": 2.903, "theoretical_loss": 4.07063744460958, "tokens_seen": 351797248 }, { "epoch": 0.11, "learning_rate": 0.0004512036108324975, "loss": 3.127, "theoretical_loss": 4.070550777049555, "tokens_seen": 351862784 }, { "epoch": 0.11, "learning_rate": 0.00045119358074222673, "loss": 3.1144, "theoretical_loss": 4.070464130149106, "tokens_seen": 351928320 }, { "epoch": 0.11, "learning_rate": 0.00045118355065195586, "loss": 2.997, "theoretical_loss": 4.07037750389946, "tokens_seen": 351993856 }, { "epoch": 0.11, "learning_rate": 0.0004511735205616851, "loss": 3.002, "theoretical_loss": 4.070290898291852, "tokens_seen": 352059392 }, { "epoch": 0.11, "learning_rate": 0.0004511634904714143, "loss": 3.0532, "theoretical_loss": 4.070204313317523, "tokens_seen": 352124928 }, { "epoch": 0.11, "learning_rate": 0.00045115346038114345, "loss": 3.0537, "theoretical_loss": 4.07011774896772, "tokens_seen": 352190464 }, { "epoch": 0.11, "objective/train/docs_used": 211560, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.582488775253296, "objective/train/theoretical_loss": 4.07003120523369, "objective/train/tokens_used": 372716000, "theoretical_loss": 4.07003120523369, "tokens_seen": 352256000 }, { "epoch": 0.11, "learning_rate": 0.00045114343029087264, "loss": 2.691, "theoretical_loss": 4.07003120523369, "tokens_seen": 352256000 }, { "epoch": 0.11, "learning_rate": 0.0004511334002006018, "loss": 3.0749, "theoretical_loss": 4.069944682106691, "tokens_seen": 352321536 }, { "epoch": 0.11, "learning_rate": 0.000451123370110331, "loss": 3.0175, "theoretical_loss": 4.0698581795779845, "tokens_seen": 352387072 }, { "epoch": 0.11, "learning_rate": 0.00045111334002006023, "loss": 3.0017, "theoretical_loss": 4.069771697638838, "tokens_seen": 352452608 }, { "epoch": 0.11, "learning_rate": 0.00045110330992978936, "loss": 3.0474, "theoretical_loss": 4.069685236280522, "tokens_seen": 352518144 }, { "epoch": 0.11, "learning_rate": 0.0004510932798395186, "loss": 2.9058, "theoretical_loss": 4.069598795494315, "tokens_seen": 352583680 }, { "epoch": 0.11, "learning_rate": 0.0004510832497492477, "loss": 2.8735, "theoretical_loss": 4.0695123752715, "tokens_seen": 352649216 }, { "epoch": 0.11, "learning_rate": 0.00045107321965897696, "loss": 2.8146, "theoretical_loss": 4.069425975603364, "tokens_seen": 352714752 }, { "epoch": 0.11, "learning_rate": 0.00045106318956870614, "loss": 2.9708, "theoretical_loss": 4.069339596481201, "tokens_seen": 352780288 }, { "epoch": 0.11, "learning_rate": 0.0004510531594784353, "loss": 2.9669, "theoretical_loss": 4.0692532378963095, "tokens_seen": 352845824 }, { "epoch": 0.11, "learning_rate": 0.0004510431293881645, "loss": 2.8697, "theoretical_loss": 4.069166899839994, "tokens_seen": 352911360 }, { "epoch": 0.11, "learning_rate": 0.00045103309929789374, "loss": 3.0089, "theoretical_loss": 4.069080582303563, "tokens_seen": 352976896 }, { "epoch": 0.11, "learning_rate": 0.00045102306920762286, "loss": 3.1818, "theoretical_loss": 4.068994285278332, "tokens_seen": 353042432 }, { "epoch": 0.11, "learning_rate": 0.0004510130391173521, "loss": 3.1709, "theoretical_loss": 4.068908008755619, "tokens_seen": 353107968 }, { "epoch": 0.11, "learning_rate": 0.0004510030090270812, "loss": 3.2865, "theoretical_loss": 4.068821752726751, "tokens_seen": 353173504 }, { "epoch": 0.11, "learning_rate": 0.00045099297893681046, "loss": 2.9514, "theoretical_loss": 4.0687355171830575, "tokens_seen": 353239040 }, { "epoch": 0.11, "learning_rate": 0.00045098294884653964, "loss": 2.7252, "theoretical_loss": 4.068649302115875, "tokens_seen": 353304576 }, { "epoch": 0.11, "learning_rate": 0.0004509729187562688, "loss": 3.2343, "theoretical_loss": 4.068563107516544, "tokens_seen": 353370112 }, { "epoch": 0.11, "learning_rate": 0.000450962888665998, "loss": 2.8777, "theoretical_loss": 4.068476933376411, "tokens_seen": 353435648 }, { "epoch": 0.11, "learning_rate": 0.0004509528585757272, "loss": 2.9855, "theoretical_loss": 4.068390779686826, "tokens_seen": 353501184 }, { "epoch": 0.11, "learning_rate": 0.00045094282848545637, "loss": 2.8759, "theoretical_loss": 4.068304646439148, "tokens_seen": 353566720 }, { "epoch": 0.11, "learning_rate": 0.0004509327983951856, "loss": 2.7545, "theoretical_loss": 4.068218533624737, "tokens_seen": 353632256 }, { "epoch": 0.11, "learning_rate": 0.00045092276830491473, "loss": 3.0406, "theoretical_loss": 4.068132441234962, "tokens_seen": 353697792 }, { "epoch": 0.11, "learning_rate": 0.00045091273821464396, "loss": 2.9013, "theoretical_loss": 4.068046369261194, "tokens_seen": 353763328 }, { "epoch": 0.11, "learning_rate": 0.00045090270812437314, "loss": 3.1276, "theoretical_loss": 4.067960317694812, "tokens_seen": 353828864 }, { "epoch": 0.11, "objective/train/docs_used": 212189, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9633750915527344, "objective/train/theoretical_loss": 4.067874286527197, "objective/train/tokens_used": 374354400, "theoretical_loss": 4.067874286527197, "tokens_seen": 353894400 }, { "epoch": 0.11, "learning_rate": 0.0004508926780341023, "loss": 3.2043, "theoretical_loss": 4.067874286527197, "tokens_seen": 353894400 }, { "epoch": 0.11, "learning_rate": 0.0004508826479438315, "loss": 2.9853, "theoretical_loss": 4.06778827574974, "tokens_seen": 353959936 }, { "epoch": 0.11, "learning_rate": 0.0004508726178535607, "loss": 3.1729, "theoretical_loss": 4.067702285353833, "tokens_seen": 354025472 }, { "epoch": 0.11, "learning_rate": 0.00045086258776328987, "loss": 3.0317, "theoretical_loss": 4.0676163153308735, "tokens_seen": 354091008 }, { "epoch": 0.11, "learning_rate": 0.0004508525576730191, "loss": 2.9049, "theoretical_loss": 4.067530365672267, "tokens_seen": 354156544 }, { "epoch": 0.11, "learning_rate": 0.00045084252758274823, "loss": 2.7865, "theoretical_loss": 4.067444436369422, "tokens_seen": 354222080 }, { "epoch": 0.11, "learning_rate": 0.00045083249749247747, "loss": 3.1346, "theoretical_loss": 4.067358527413753, "tokens_seen": 354287616 }, { "epoch": 0.11, "learning_rate": 0.0004508224674022066, "loss": 2.8867, "theoretical_loss": 4.06727263879668, "tokens_seen": 354353152 }, { "epoch": 0.11, "learning_rate": 0.00045081243731193583, "loss": 2.9152, "theoretical_loss": 4.067186770509627, "tokens_seen": 354418688 }, { "epoch": 0.11, "learning_rate": 0.000450802407221665, "loss": 2.8679, "theoretical_loss": 4.067100922544023, "tokens_seen": 354484224 }, { "epoch": 0.11, "learning_rate": 0.0004507923771313942, "loss": 2.7877, "theoretical_loss": 4.067015094891306, "tokens_seen": 354549760 }, { "epoch": 0.11, "learning_rate": 0.00045078234704112337, "loss": 2.9082, "theoretical_loss": 4.066929287542914, "tokens_seen": 354615296 }, { "epoch": 0.11, "learning_rate": 0.00045077231695085255, "loss": 3.1084, "theoretical_loss": 4.066843500490292, "tokens_seen": 354680832 }, { "epoch": 0.11, "learning_rate": 0.00045076228686058173, "loss": 2.8425, "theoretical_loss": 4.066757733724893, "tokens_seen": 354746368 }, { "epoch": 0.11, "learning_rate": 0.00045075225677031097, "loss": 3.1105, "theoretical_loss": 4.066671987238171, "tokens_seen": 354811904 }, { "epoch": 0.11, "learning_rate": 0.0004507422266800401, "loss": 2.7815, "theoretical_loss": 4.0665862610215875, "tokens_seen": 354877440 }, { "epoch": 0.11, "learning_rate": 0.00045073219658976933, "loss": 2.8093, "theoretical_loss": 4.066500555066609, "tokens_seen": 354942976 }, { "epoch": 0.11, "learning_rate": 0.0004507221664994985, "loss": 3.1331, "theoretical_loss": 4.066414869364707, "tokens_seen": 355008512 }, { "epoch": 0.11, "learning_rate": 0.0004507121364092277, "loss": 3.0502, "theoretical_loss": 4.066329203907358, "tokens_seen": 355074048 }, { "epoch": 0.11, "learning_rate": 0.0004507021063189569, "loss": 2.9057, "theoretical_loss": 4.066243558686043, "tokens_seen": 355139584 }, { "epoch": 0.11, "learning_rate": 0.00045069207622868606, "loss": 2.8425, "theoretical_loss": 4.066157933692249, "tokens_seen": 355205120 }, { "epoch": 0.11, "learning_rate": 0.00045068204613841524, "loss": 2.8221, "theoretical_loss": 4.066072328917469, "tokens_seen": 355270656 }, { "epoch": 0.11, "learning_rate": 0.0004506720160481445, "loss": 2.9991, "theoretical_loss": 4.0659867443532, "tokens_seen": 355336192 }, { "epoch": 0.11, "learning_rate": 0.0004506619859578736, "loss": 3.0212, "theoretical_loss": 4.065901179990943, "tokens_seen": 355401728 }, { "epoch": 0.11, "learning_rate": 0.00045065195586760284, "loss": 2.8883, "theoretical_loss": 4.065815635822206, "tokens_seen": 355467264 }, { "epoch": 0.11, "objective/train/docs_used": 213521, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.758910894393921, "objective/train/theoretical_loss": 4.065730111838502, "objective/train/tokens_used": 375992800, "theoretical_loss": 4.065730111838502, "tokens_seen": 355532800 }, { "epoch": 0.11, "learning_rate": 0.00045064192577733196, "loss": 3.0933, "theoretical_loss": 4.065730111838502, "tokens_seen": 355532800 }, { "epoch": 0.11, "learning_rate": 0.0004506318956870612, "loss": 3.0034, "theoretical_loss": 4.0656446080313495, "tokens_seen": 355598336 }, { "epoch": 0.11, "learning_rate": 0.0004506218655967904, "loss": 2.9333, "theoretical_loss": 4.065559124392269, "tokens_seen": 355663872 }, { "epoch": 0.11, "learning_rate": 0.00045061183550651956, "loss": 3.1183, "theoretical_loss": 4.065473660912791, "tokens_seen": 355729408 }, { "epoch": 0.11, "learning_rate": 0.00045060180541624874, "loss": 3.0313, "theoretical_loss": 4.065388217584447, "tokens_seen": 355794944 }, { "epoch": 0.11, "learning_rate": 0.0004505917753259779, "loss": 2.7215, "theoretical_loss": 4.065302794398775, "tokens_seen": 355860480 }, { "epoch": 0.11, "learning_rate": 0.0004505817452357071, "loss": 2.8688, "theoretical_loss": 4.0652173913473195, "tokens_seen": 355926016 }, { "epoch": 0.11, "learning_rate": 0.00045057171514543634, "loss": 2.7887, "theoretical_loss": 4.065132008421628, "tokens_seen": 355991552 }, { "epoch": 0.11, "learning_rate": 0.00045056168505516547, "loss": 2.9822, "theoretical_loss": 4.065046645613254, "tokens_seen": 356057088 }, { "epoch": 0.11, "learning_rate": 0.0004505516549648947, "loss": 3.0194, "theoretical_loss": 4.0649613029137575, "tokens_seen": 356122624 }, { "epoch": 0.11, "learning_rate": 0.0004505416248746239, "loss": 2.9109, "theoretical_loss": 4.0648759803147, "tokens_seen": 356188160 }, { "epoch": 0.11, "learning_rate": 0.00045053159478435306, "loss": 2.8684, "theoretical_loss": 4.064790677807652, "tokens_seen": 356253696 }, { "epoch": 0.11, "learning_rate": 0.00045052156469408224, "loss": 2.7241, "theoretical_loss": 4.064705395384187, "tokens_seen": 356319232 }, { "epoch": 0.11, "learning_rate": 0.0004505115346038114, "loss": 2.9499, "theoretical_loss": 4.064620133035883, "tokens_seen": 356384768 }, { "epoch": 0.11, "learning_rate": 0.0004505015045135406, "loss": 2.9387, "theoretical_loss": 4.0645348907543255, "tokens_seen": 356450304 }, { "epoch": 0.11, "learning_rate": 0.00045049147442326984, "loss": 2.9353, "theoretical_loss": 4.064449668531103, "tokens_seen": 356515840 }, { "epoch": 0.11, "learning_rate": 0.00045048144433299897, "loss": 2.7079, "theoretical_loss": 4.064364466357809, "tokens_seen": 356581376 }, { "epoch": 0.11, "learning_rate": 0.0004504714142427282, "loss": 2.8404, "theoretical_loss": 4.064279284226044, "tokens_seen": 356646912 }, { "epoch": 0.11, "learning_rate": 0.00045046138415245733, "loss": 2.9141, "theoretical_loss": 4.064194122127411, "tokens_seen": 356712448 }, { "epoch": 0.11, "learning_rate": 0.00045045135406218657, "loss": 2.9891, "theoretical_loss": 4.06410898005352, "tokens_seen": 356777984 }, { "epoch": 0.11, "learning_rate": 0.0004504413239719158, "loss": 2.9367, "theoretical_loss": 4.064023857995986, "tokens_seen": 356843520 }, { "epoch": 0.11, "learning_rate": 0.00045043129388164493, "loss": 2.8608, "theoretical_loss": 4.0639387559464275, "tokens_seen": 356909056 }, { "epoch": 0.11, "learning_rate": 0.00045042126379137416, "loss": 2.9035, "theoretical_loss": 4.063853673896469, "tokens_seen": 356974592 }, { "epoch": 0.11, "learning_rate": 0.00045041123370110334, "loss": 2.9264, "theoretical_loss": 4.063768611837742, "tokens_seen": 357040128 }, { "epoch": 0.11, "learning_rate": 0.0004504012036108325, "loss": 2.9756, "theoretical_loss": 4.063683569761879, "tokens_seen": 357105664 }, { "epoch": 0.11, "objective/train/docs_used": 214188, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7816708087921143, "objective/train/theoretical_loss": 4.063598547660519, "objective/train/tokens_used": 377631200, "theoretical_loss": 4.063598547660519, "tokens_seen": 357171200 }, { "epoch": 0.11, "learning_rate": 0.0004503911735205617, "loss": 2.8588, "theoretical_loss": 4.063598547660519, "tokens_seen": 357171200 }, { "epoch": 0.11, "learning_rate": 0.0004503811434302909, "loss": 2.9712, "theoretical_loss": 4.06351354552531, "tokens_seen": 357236736 }, { "epoch": 0.11, "learning_rate": 0.00045037111334002007, "loss": 2.9797, "theoretical_loss": 4.063428563347898, "tokens_seen": 357302272 }, { "epoch": 0.11, "learning_rate": 0.0004503610832497493, "loss": 2.8845, "theoretical_loss": 4.063343601119941, "tokens_seen": 357367808 }, { "epoch": 0.11, "learning_rate": 0.00045035105315947843, "loss": 2.6153, "theoretical_loss": 4.063258658833096, "tokens_seen": 357433344 }, { "epoch": 0.11, "learning_rate": 0.00045034102306920767, "loss": 2.855, "theoretical_loss": 4.06317373647903, "tokens_seen": 357498880 }, { "epoch": 0.11, "learning_rate": 0.0004503309929789368, "loss": 2.9601, "theoretical_loss": 4.0630888340494105, "tokens_seen": 357564416 }, { "epoch": 0.11, "learning_rate": 0.00045032096288866603, "loss": 3.0078, "theoretical_loss": 4.063003951535913, "tokens_seen": 357629952 }, { "epoch": 0.11, "learning_rate": 0.0004503109327983952, "loss": 2.8686, "theoretical_loss": 4.06291908893022, "tokens_seen": 357695488 }, { "epoch": 0.11, "learning_rate": 0.0004503009027081244, "loss": 2.9919, "theoretical_loss": 4.062834246224012, "tokens_seen": 357761024 }, { "epoch": 0.11, "learning_rate": 0.00045029087261785357, "loss": 2.882, "theoretical_loss": 4.062749423408981, "tokens_seen": 357826560 }, { "epoch": 0.11, "learning_rate": 0.00045028084252758275, "loss": 2.8618, "theoretical_loss": 4.062664620476822, "tokens_seen": 357892096 }, { "epoch": 0.11, "learning_rate": 0.00045027081243731193, "loss": 2.8418, "theoretical_loss": 4.062579837419235, "tokens_seen": 357957632 }, { "epoch": 0.11, "learning_rate": 0.00045026078234704117, "loss": 2.8871, "theoretical_loss": 4.062495074227923, "tokens_seen": 358023168 }, { "epoch": 0.11, "learning_rate": 0.0004502507522567703, "loss": 3.189, "theoretical_loss": 4.062410330894597, "tokens_seen": 358088704 }, { "epoch": 0.11, "learning_rate": 0.00045024072216649953, "loss": 2.8227, "theoretical_loss": 4.062325607410972, "tokens_seen": 358154240 }, { "epoch": 0.11, "learning_rate": 0.0004502306920762287, "loss": 3.106, "theoretical_loss": 4.062240903768767, "tokens_seen": 358219776 }, { "epoch": 0.11, "learning_rate": 0.0004502206619859579, "loss": 2.8611, "theoretical_loss": 4.062156219959705, "tokens_seen": 358285312 }, { "epoch": 0.11, "learning_rate": 0.0004502106318956871, "loss": 2.7625, "theoretical_loss": 4.062071555975519, "tokens_seen": 358350848 }, { "epoch": 0.11, "learning_rate": 0.00045020060180541626, "loss": 2.8433, "theoretical_loss": 4.061986911807942, "tokens_seen": 358416384 }, { "epoch": 0.11, "learning_rate": 0.00045019057171514544, "loss": 2.9341, "theoretical_loss": 4.061902287448714, "tokens_seen": 358481920 }, { "epoch": 0.11, "learning_rate": 0.0004501805416248747, "loss": 2.9221, "theoretical_loss": 4.061817682889577, "tokens_seen": 358547456 }, { "epoch": 0.11, "learning_rate": 0.0004501705115346038, "loss": 3.0211, "theoretical_loss": 4.061733098122284, "tokens_seen": 358612992 }, { "epoch": 0.11, "learning_rate": 0.00045016048144433304, "loss": 2.7245, "theoretical_loss": 4.061648533138587, "tokens_seen": 358678528 }, { "epoch": 0.11, "learning_rate": 0.00045015045135406216, "loss": 2.7328, "theoretical_loss": 4.0615639879302465, "tokens_seen": 358744064 }, { "epoch": 0.11, "objective/train/docs_used": 215284, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.302658796310425, "objective/train/theoretical_loss": 4.061479462489026, "objective/train/tokens_used": 379269600, "theoretical_loss": 4.061479462489026, "tokens_seen": 358809600 }, { "epoch": 0.11, "learning_rate": 0.0004501404212637914, "loss": 2.824, "theoretical_loss": 4.061479462489026, "tokens_seen": 358809600 }, { "epoch": 0.11, "learning_rate": 0.0004501303911735206, "loss": 2.6842, "theoretical_loss": 4.061394956806694, "tokens_seen": 358875136 }, { "epoch": 0.11, "learning_rate": 0.00045012036108324976, "loss": 2.8211, "theoretical_loss": 4.061310470875027, "tokens_seen": 358940672 }, { "epoch": 0.11, "learning_rate": 0.00045011033099297894, "loss": 2.9749, "theoretical_loss": 4.0612260046858015, "tokens_seen": 359006208 }, { "epoch": 0.11, "learning_rate": 0.0004501003009027081, "loss": 3.2034, "theoretical_loss": 4.061141558230803, "tokens_seen": 359071744 }, { "epoch": 0.11, "learning_rate": 0.0004500902708124373, "loss": 2.9755, "theoretical_loss": 4.06105713150182, "tokens_seen": 359137280 }, { "epoch": 0.11, "learning_rate": 0.00045008024072216654, "loss": 2.9855, "theoretical_loss": 4.060972724490646, "tokens_seen": 359202816 }, { "epoch": 0.11, "learning_rate": 0.00045007021063189567, "loss": 3.1566, "theoretical_loss": 4.0608883371890805, "tokens_seen": 359268352 }, { "epoch": 0.11, "learning_rate": 0.0004500601805416249, "loss": 3.1019, "theoretical_loss": 4.060803969588925, "tokens_seen": 359333888 }, { "epoch": 0.11, "learning_rate": 0.0004500501504513541, "loss": 3.2989, "theoretical_loss": 4.06071962168199, "tokens_seen": 359399424 }, { "epoch": 0.11, "learning_rate": 0.00045004012036108326, "loss": 2.6658, "theoretical_loss": 4.060635293460089, "tokens_seen": 359464960 }, { "epoch": 0.11, "learning_rate": 0.00045003009027081244, "loss": 2.9667, "theoretical_loss": 4.060550984915039, "tokens_seen": 359530496 }, { "epoch": 0.11, "learning_rate": 0.0004500200601805416, "loss": 2.8963, "theoretical_loss": 4.060466696038664, "tokens_seen": 359596032 }, { "epoch": 0.11, "learning_rate": 0.0004500100300902708, "loss": 2.9883, "theoretical_loss": 4.060382426822792, "tokens_seen": 359661568 }, { "epoch": 0.11, "learning_rate": 0.00045000000000000004, "loss": 2.9093, "theoretical_loss": 4.060298177259256, "tokens_seen": 359727104 }, { "epoch": 0.11, "learning_rate": 0.00044998996990972917, "loss": 3.025, "theoretical_loss": 4.060213947339894, "tokens_seen": 359792640 }, { "epoch": 0.11, "learning_rate": 0.0004499799398194584, "loss": 3.1524, "theoretical_loss": 4.060129737056547, "tokens_seen": 359858176 }, { "epoch": 0.11, "learning_rate": 0.00044996990972918753, "loss": 2.7814, "theoretical_loss": 4.060045546401065, "tokens_seen": 359923712 }, { "epoch": 0.11, "learning_rate": 0.00044995987963891677, "loss": 3.1582, "theoretical_loss": 4.0599613753653, "tokens_seen": 359989248 }, { "epoch": 0.11, "learning_rate": 0.00044994984954864595, "loss": 2.9273, "theoretical_loss": 4.0598772239411085, "tokens_seen": 360054784 }, { "epoch": 0.11, "learning_rate": 0.00044993981945837513, "loss": 3.3608, "theoretical_loss": 4.059793092120353, "tokens_seen": 360120320 }, { "epoch": 0.11, "learning_rate": 0.0004499297893681043, "loss": 2.8893, "theoretical_loss": 4.0597089798949, "tokens_seen": 360185856 }, { "epoch": 0.11, "learning_rate": 0.00044991975927783355, "loss": 2.9264, "theoretical_loss": 4.059624887256622, "tokens_seen": 360251392 }, { "epoch": 0.11, "learning_rate": 0.00044990972918756267, "loss": 2.929, "theoretical_loss": 4.059540814197396, "tokens_seen": 360316928 }, { "epoch": 0.11, "learning_rate": 0.0004498996990972919, "loss": 3.0329, "theoretical_loss": 4.059456760709104, "tokens_seen": 360382464 }, { "epoch": 0.11, "objective/train/docs_used": 215704, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.066336154937744, "objective/train/theoretical_loss": 4.05937272678363, "objective/train/tokens_used": 380908000, "theoretical_loss": 4.05937272678363, "tokens_seen": 360448000 }, { "epoch": 0.11, "learning_rate": 0.00044988966900702103, "loss": 2.8475, "theoretical_loss": 4.05937272678363, "tokens_seen": 360448000 }, { "epoch": 0.11, "learning_rate": 0.00044987963891675027, "loss": 2.8076, "theoretical_loss": 4.059288712412868, "tokens_seen": 360513536 }, { "epoch": 0.11, "learning_rate": 0.00044986960882647945, "loss": 2.908, "theoretical_loss": 4.059204717588712, "tokens_seen": 360579072 }, { "epoch": 0.11, "learning_rate": 0.00044985957873620863, "loss": 3.1984, "theoretical_loss": 4.059120742303065, "tokens_seen": 360644608 }, { "epoch": 0.11, "learning_rate": 0.0004498495486459378, "loss": 2.9602, "theoretical_loss": 4.059036786547831, "tokens_seen": 360710144 }, { "epoch": 0.11, "learning_rate": 0.000449839518555667, "loss": 3.3413, "theoretical_loss": 4.058952850314922, "tokens_seen": 360775680 }, { "epoch": 0.11, "learning_rate": 0.0004498294884653962, "loss": 2.8901, "theoretical_loss": 4.058868933596252, "tokens_seen": 360841216 }, { "epoch": 0.11, "learning_rate": 0.0004498194583751254, "loss": 2.9428, "theoretical_loss": 4.0587850363837426, "tokens_seen": 360906752 }, { "epoch": 0.11, "learning_rate": 0.00044980942828485454, "loss": 2.9984, "theoretical_loss": 4.058701158669318, "tokens_seen": 360972288 }, { "epoch": 0.11, "learning_rate": 0.00044979939819458377, "loss": 3.1567, "theoretical_loss": 4.058617300444908, "tokens_seen": 361037824 }, { "epoch": 0.11, "learning_rate": 0.0004497893681043129, "loss": 2.8662, "theoretical_loss": 4.0585334617024476, "tokens_seen": 361103360 }, { "epoch": 0.11, "learning_rate": 0.00044977933801404213, "loss": 2.9641, "theoretical_loss": 4.058449642433875, "tokens_seen": 361168896 }, { "epoch": 0.11, "learning_rate": 0.0004497693079237713, "loss": 2.7841, "theoretical_loss": 4.058365842631137, "tokens_seen": 361234432 }, { "epoch": 0.11, "learning_rate": 0.0004497592778335005, "loss": 2.808, "theoretical_loss": 4.05828206228618, "tokens_seen": 361299968 }, { "epoch": 0.11, "learning_rate": 0.0004497492477432297, "loss": 2.8638, "theoretical_loss": 4.058198301390959, "tokens_seen": 361365504 }, { "epoch": 0.11, "learning_rate": 0.0004497392176529589, "loss": 2.7483, "theoretical_loss": 4.058114559937433, "tokens_seen": 361431040 }, { "epoch": 0.11, "learning_rate": 0.00044972918756268804, "loss": 2.8876, "theoretical_loss": 4.058030837917565, "tokens_seen": 361496576 }, { "epoch": 0.11, "learning_rate": 0.0004497191574724173, "loss": 2.8564, "theoretical_loss": 4.057947135323323, "tokens_seen": 361562112 }, { "epoch": 0.11, "learning_rate": 0.0004497091273821464, "loss": 3.2562, "theoretical_loss": 4.057863452146679, "tokens_seen": 361627648 }, { "epoch": 0.11, "learning_rate": 0.00044969909729187564, "loss": 2.8309, "theoretical_loss": 4.057779788379613, "tokens_seen": 361693184 }, { "epoch": 0.11, "learning_rate": 0.0004496890672016049, "loss": 2.8252, "theoretical_loss": 4.057696144014106, "tokens_seen": 361758720 }, { "epoch": 0.11, "learning_rate": 0.000449679037111334, "loss": 3.1268, "theoretical_loss": 4.057612519042145, "tokens_seen": 361824256 }, { "epoch": 0.11, "learning_rate": 0.00044966900702106324, "loss": 2.7978, "theoretical_loss": 4.057528913455723, "tokens_seen": 361889792 }, { "epoch": 0.11, "learning_rate": 0.00044965897693079236, "loss": 2.8384, "theoretical_loss": 4.057445327246836, "tokens_seen": 361955328 }, { "epoch": 0.11, "learning_rate": 0.0004496489468405216, "loss": 2.6464, "theoretical_loss": 4.0573617604074865, "tokens_seen": 362020864 }, { "epoch": 0.11, "objective/train/docs_used": 216313, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2176437377929688, "objective/train/theoretical_loss": 4.0572782129296785, "objective/train/tokens_used": 382546400, "theoretical_loss": 4.0572782129296785, "tokens_seen": 362086400 }, { "epoch": 0.11, "learning_rate": 0.0004496389167502508, "loss": 3.0372, "theoretical_loss": 4.0572782129296785, "tokens_seen": 362086400 }, { "epoch": 0.11, "learning_rate": 0.00044962888665997996, "loss": 2.9452, "theoretical_loss": 4.057194684805426, "tokens_seen": 362151936 }, { "epoch": 0.11, "learning_rate": 0.00044961885656970914, "loss": 3.0605, "theoretical_loss": 4.057111176026742, "tokens_seen": 362217472 }, { "epoch": 0.11, "learning_rate": 0.0004496088264794383, "loss": 2.9695, "theoretical_loss": 4.0570276865856485, "tokens_seen": 362283008 }, { "epoch": 0.11, "learning_rate": 0.0004495987963891675, "loss": 2.8051, "theoretical_loss": 4.056944216474171, "tokens_seen": 362348544 }, { "epoch": 0.11, "learning_rate": 0.00044958876629889674, "loss": 3.0178, "theoretical_loss": 4.056860765684338, "tokens_seen": 362414080 }, { "epoch": 0.11, "learning_rate": 0.00044957873620862587, "loss": 3.0671, "theoretical_loss": 4.056777334208186, "tokens_seen": 362479616 }, { "epoch": 0.11, "learning_rate": 0.0004495687061183551, "loss": 2.9791, "theoretical_loss": 4.056693922037753, "tokens_seen": 362545152 }, { "epoch": 0.11, "learning_rate": 0.0004495586760280843, "loss": 2.8048, "theoretical_loss": 4.056610529165083, "tokens_seen": 362610688 }, { "epoch": 0.11, "learning_rate": 0.00044954864593781346, "loss": 3.1233, "theoretical_loss": 4.056527155582226, "tokens_seen": 362676224 }, { "epoch": 0.11, "learning_rate": 0.00044953861584754264, "loss": 2.8641, "theoretical_loss": 4.056443801281235, "tokens_seen": 362741760 }, { "epoch": 0.11, "learning_rate": 0.0004495285857572718, "loss": 2.8788, "theoretical_loss": 4.056360466254168, "tokens_seen": 362807296 }, { "epoch": 0.11, "learning_rate": 0.000449518555667001, "loss": 2.911, "theoretical_loss": 4.056277150493088, "tokens_seen": 362872832 }, { "epoch": 0.11, "learning_rate": 0.00044950852557673024, "loss": 2.8492, "theoretical_loss": 4.056193853990063, "tokens_seen": 362938368 }, { "epoch": 0.11, "learning_rate": 0.00044949849548645937, "loss": 3.0162, "theoretical_loss": 4.056110576737166, "tokens_seen": 363003904 }, { "epoch": 0.11, "learning_rate": 0.0004494884653961886, "loss": 3.2349, "theoretical_loss": 4.056027318726472, "tokens_seen": 363069440 }, { "epoch": 0.11, "learning_rate": 0.00044947843530591773, "loss": 2.8015, "theoretical_loss": 4.055944079950065, "tokens_seen": 363134976 }, { "epoch": 0.11, "learning_rate": 0.00044946840521564697, "loss": 2.9836, "theoretical_loss": 4.0558608604000295, "tokens_seen": 363200512 }, { "epoch": 0.11, "learning_rate": 0.00044945837512537615, "loss": 2.956, "theoretical_loss": 4.055777660068458, "tokens_seen": 363266048 }, { "epoch": 0.11, "learning_rate": 0.00044944834503510533, "loss": 2.9963, "theoretical_loss": 4.055694478947445, "tokens_seen": 363331584 }, { "epoch": 0.11, "learning_rate": 0.0004494383149448345, "loss": 2.8188, "theoretical_loss": 4.055611317029093, "tokens_seen": 363397120 }, { "epoch": 0.11, "learning_rate": 0.00044942828485456375, "loss": 2.9132, "theoretical_loss": 4.055528174305505, "tokens_seen": 363462656 }, { "epoch": 0.11, "learning_rate": 0.00044941825476429287, "loss": 2.8309, "theoretical_loss": 4.055445050768792, "tokens_seen": 363528192 }, { "epoch": 0.11, "learning_rate": 0.0004494082246740221, "loss": 3.0976, "theoretical_loss": 4.055361946411068, "tokens_seen": 363593728 }, { "epoch": 0.11, "learning_rate": 0.00044939819458375123, "loss": 2.7774, "theoretical_loss": 4.055278861224452, "tokens_seen": 363659264 }, { "epoch": 0.11, "objective/train/docs_used": 217479, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.15562105178833, "objective/train/theoretical_loss": 4.055195795201069, "objective/train/tokens_used": 384184800, "theoretical_loss": 4.055195795201069, "tokens_seen": 363724800 }, { "epoch": 0.11, "learning_rate": 0.00044938816449348047, "loss": 2.9052, "theoretical_loss": 4.055195795201069, "tokens_seen": 363724800 }, { "epoch": 0.11, "learning_rate": 0.00044937813440320965, "loss": 3.0289, "theoretical_loss": 4.055112748333046, "tokens_seen": 363790336 }, { "epoch": 0.11, "learning_rate": 0.00044936810431293883, "loss": 3.0627, "theoretical_loss": 4.0550297206125165, "tokens_seen": 363855872 }, { "epoch": 0.11, "learning_rate": 0.000449358074222668, "loss": 2.8629, "theoretical_loss": 4.054946712031619, "tokens_seen": 363921408 }, { "epoch": 0.11, "learning_rate": 0.0004493480441323972, "loss": 3.1226, "theoretical_loss": 4.054863722582494, "tokens_seen": 363986944 }, { "epoch": 0.11, "learning_rate": 0.0004493380140421264, "loss": 2.8903, "theoretical_loss": 4.054780752257291, "tokens_seen": 364052480 }, { "epoch": 0.11, "learning_rate": 0.0004493279839518556, "loss": 2.7772, "theoretical_loss": 4.05469780104816, "tokens_seen": 364118016 }, { "epoch": 0.11, "learning_rate": 0.00044931795386158474, "loss": 2.8789, "theoretical_loss": 4.054614868947258, "tokens_seen": 364183552 }, { "epoch": 0.11, "learning_rate": 0.00044930792377131397, "loss": 2.9326, "theoretical_loss": 4.054531955946745, "tokens_seen": 364249088 }, { "epoch": 0.11, "learning_rate": 0.0004492978936810431, "loss": 2.9398, "theoretical_loss": 4.054449062038787, "tokens_seen": 364314624 }, { "epoch": 0.11, "learning_rate": 0.00044928786359077234, "loss": 3.1035, "theoretical_loss": 4.054366187215556, "tokens_seen": 364380160 }, { "epoch": 0.11, "learning_rate": 0.0004492778335005015, "loss": 2.9538, "theoretical_loss": 4.0542833314692235, "tokens_seen": 364445696 }, { "epoch": 0.11, "learning_rate": 0.0004492678034102307, "loss": 2.7926, "theoretical_loss": 4.054200494791972, "tokens_seen": 364511232 }, { "epoch": 0.11, "learning_rate": 0.0004492577733199599, "loss": 3.1084, "theoretical_loss": 4.054117677175983, "tokens_seen": 364576768 }, { "epoch": 0.11, "learning_rate": 0.0004492477432296891, "loss": 2.8922, "theoretical_loss": 4.054034878613447, "tokens_seen": 364642304 }, { "epoch": 0.11, "learning_rate": 0.00044923771313941824, "loss": 3.0288, "theoretical_loss": 4.053952099096556, "tokens_seen": 364707840 }, { "epoch": 0.11, "learning_rate": 0.0004492276830491475, "loss": 3.0087, "theoretical_loss": 4.053869338617508, "tokens_seen": 364773376 }, { "epoch": 0.11, "learning_rate": 0.0004492176529588766, "loss": 2.8053, "theoretical_loss": 4.053786597168506, "tokens_seen": 364838912 }, { "epoch": 0.11, "learning_rate": 0.00044920762286860584, "loss": 3.0586, "theoretical_loss": 4.053703874741757, "tokens_seen": 364904448 }, { "epoch": 0.11, "learning_rate": 0.000449197592778335, "loss": 2.9923, "theoretical_loss": 4.053621171329472, "tokens_seen": 364969984 }, { "epoch": 0.11, "learning_rate": 0.0004491875626880642, "loss": 3.195, "theoretical_loss": 4.053538486923868, "tokens_seen": 365035520 }, { "epoch": 0.11, "learning_rate": 0.0004491775325977934, "loss": 3.128, "theoretical_loss": 4.053455821517165, "tokens_seen": 365101056 }, { "epoch": 0.11, "learning_rate": 0.00044916750250752256, "loss": 3.0488, "theoretical_loss": 4.053373175101589, "tokens_seen": 365166592 }, { "epoch": 0.11, "learning_rate": 0.00044915747241725174, "loss": 3.1723, "theoretical_loss": 4.053290547669369, "tokens_seen": 365232128 }, { "epoch": 0.11, "learning_rate": 0.000449147442326981, "loss": 3.107, "theoretical_loss": 4.05320793921274, "tokens_seen": 365297664 }, { "epoch": 0.11, "objective/train/docs_used": 218231, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4497435092926025, "objective/train/theoretical_loss": 4.053125349723942, "objective/train/tokens_used": 385823200, "theoretical_loss": 4.053125349723942, "tokens_seen": 365363200 }, { "epoch": 0.11, "learning_rate": 0.0004491374122367101, "loss": 2.7664, "theoretical_loss": 4.053125349723942, "tokens_seen": 365363200 }, { "epoch": 0.11, "learning_rate": 0.00044912738214643934, "loss": 3.0456, "theoretical_loss": 4.053042779195217, "tokens_seen": 365428736 }, { "epoch": 0.11, "learning_rate": 0.00044911735205616847, "loss": 2.9308, "theoretical_loss": 4.052960227618815, "tokens_seen": 365494272 }, { "epoch": 0.11, "learning_rate": 0.0004491073219658977, "loss": 3.0695, "theoretical_loss": 4.052877694986987, "tokens_seen": 365559808 }, { "epoch": 0.11, "learning_rate": 0.0004490972918756269, "loss": 3.0834, "theoretical_loss": 4.052795181291992, "tokens_seen": 365625344 }, { "epoch": 0.11, "learning_rate": 0.00044908726178535607, "loss": 3.0867, "theoretical_loss": 4.05271268652609, "tokens_seen": 365690880 }, { "epoch": 0.11, "learning_rate": 0.00044907723169508525, "loss": 3.1497, "theoretical_loss": 4.052630210681549, "tokens_seen": 365756416 }, { "epoch": 0.11, "learning_rate": 0.0004490672016048145, "loss": 2.7121, "theoretical_loss": 4.052547753750639, "tokens_seen": 365821952 }, { "epoch": 0.11, "learning_rate": 0.0004490571715145436, "loss": 3.0036, "theoretical_loss": 4.0524653157256365, "tokens_seen": 365887488 }, { "epoch": 0.11, "learning_rate": 0.00044904714142427284, "loss": 2.9278, "theoretical_loss": 4.05238289659882, "tokens_seen": 365953024 }, { "epoch": 0.11, "learning_rate": 0.00044903711133400197, "loss": 2.7647, "theoretical_loss": 4.052300496362475, "tokens_seen": 366018560 }, { "epoch": 0.11, "learning_rate": 0.0004490270812437312, "loss": 2.772, "theoretical_loss": 4.052218115008892, "tokens_seen": 366084096 }, { "epoch": 0.11, "learning_rate": 0.0004490170511534604, "loss": 2.921, "theoretical_loss": 4.052135752530362, "tokens_seen": 366149632 }, { "epoch": 0.11, "learning_rate": 0.00044900702106318957, "loss": 3.1039, "theoretical_loss": 4.0520534089191855, "tokens_seen": 366215168 }, { "epoch": 0.11, "learning_rate": 0.00044899699097291875, "loss": 2.7293, "theoretical_loss": 4.051971084167663, "tokens_seen": 366280704 }, { "epoch": 0.11, "learning_rate": 0.00044898696088264793, "loss": 2.949, "theoretical_loss": 4.051888778268103, "tokens_seen": 366346240 }, { "epoch": 0.11, "learning_rate": 0.0004489769307923771, "loss": 3.0874, "theoretical_loss": 4.051806491212817, "tokens_seen": 366411776 }, { "epoch": 0.11, "learning_rate": 0.00044896690070210635, "loss": 2.9753, "theoretical_loss": 4.051724222994122, "tokens_seen": 366477312 }, { "epoch": 0.11, "learning_rate": 0.0004489568706118355, "loss": 3.0817, "theoretical_loss": 4.051641973604337, "tokens_seen": 366542848 }, { "epoch": 0.11, "learning_rate": 0.0004489468405215647, "loss": 2.9177, "theoretical_loss": 4.051559743035789, "tokens_seen": 366608384 }, { "epoch": 0.11, "learning_rate": 0.00044893681043129395, "loss": 3.1013, "theoretical_loss": 4.051477531280806, "tokens_seen": 366673920 }, { "epoch": 0.11, "learning_rate": 0.00044892678034102307, "loss": 2.8753, "theoretical_loss": 4.051395338331723, "tokens_seen": 366739456 }, { "epoch": 0.11, "learning_rate": 0.0004489167502507523, "loss": 3.0245, "theoretical_loss": 4.05131316418088, "tokens_seen": 366804992 }, { "epoch": 0.11, "learning_rate": 0.00044890672016048143, "loss": 2.9561, "theoretical_loss": 4.051231008820618, "tokens_seen": 366870528 }, { "epoch": 0.11, "learning_rate": 0.00044889669007021067, "loss": 3.0036, "theoretical_loss": 4.051148872243286, "tokens_seen": 366936064 }, { "epoch": 0.11, "objective/train/docs_used": 219467, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0995004177093506, "objective/train/theoretical_loss": 4.051066754441235, "objective/train/tokens_used": 387461600, "theoretical_loss": 4.051066754441235, "tokens_seen": 367001600 }, { "epoch": 0.11, "learning_rate": 0.00044888665997993985, "loss": 3.0026, "theoretical_loss": 4.051066754441235, "tokens_seen": 367001600 }, { "epoch": 0.11, "learning_rate": 0.00044887662988966903, "loss": 3.1344, "theoretical_loss": 4.050984655406824, "tokens_seen": 367067136 }, { "epoch": 0.11, "learning_rate": 0.0004488665997993982, "loss": 3.0819, "theoretical_loss": 4.050902575132412, "tokens_seen": 367132672 }, { "epoch": 0.11, "learning_rate": 0.0004488565697091274, "loss": 3.0175, "theoretical_loss": 4.050820513610366, "tokens_seen": 367198208 }, { "epoch": 0.11, "learning_rate": 0.0004488465396188566, "loss": 2.9302, "theoretical_loss": 4.050738470833054, "tokens_seen": 367263744 }, { "epoch": 0.11, "learning_rate": 0.0004488365095285858, "loss": 3.0906, "theoretical_loss": 4.0506564467928525, "tokens_seen": 367329280 }, { "epoch": 0.11, "learning_rate": 0.00044882647943831494, "loss": 2.7848, "theoretical_loss": 4.05057444148214, "tokens_seen": 367394816 }, { "epoch": 0.11, "learning_rate": 0.0004488164493480442, "loss": 2.9642, "theoretical_loss": 4.0504924548933, "tokens_seen": 367460352 }, { "epoch": 0.11, "learning_rate": 0.0004488064192577733, "loss": 2.9148, "theoretical_loss": 4.05041048701872, "tokens_seen": 367525888 }, { "epoch": 0.11, "learning_rate": 0.00044879638916750254, "loss": 2.9993, "theoretical_loss": 4.050328537850792, "tokens_seen": 367591424 }, { "epoch": 0.11, "learning_rate": 0.0004487863590772317, "loss": 3.2353, "theoretical_loss": 4.050246607381913, "tokens_seen": 367656960 }, { "epoch": 0.11, "learning_rate": 0.0004487763289869609, "loss": 3.096, "theoretical_loss": 4.0501646956044866, "tokens_seen": 367722496 }, { "epoch": 0.11, "learning_rate": 0.0004487662988966901, "loss": 2.7814, "theoretical_loss": 4.050082802510915, "tokens_seen": 367788032 }, { "epoch": 0.11, "learning_rate": 0.0004487562688064193, "loss": 3.0983, "theoretical_loss": 4.050000928093611, "tokens_seen": 367853568 }, { "epoch": 0.11, "learning_rate": 0.00044874623871614844, "loss": 3.2659, "theoretical_loss": 4.049919072344987, "tokens_seen": 367919104 }, { "epoch": 0.11, "learning_rate": 0.0004487362086258777, "loss": 3.0178, "theoretical_loss": 4.049837235257463, "tokens_seen": 367984640 }, { "epoch": 0.11, "learning_rate": 0.0004487261785356068, "loss": 2.9854, "theoretical_loss": 4.049755416823463, "tokens_seen": 368050176 }, { "epoch": 0.11, "learning_rate": 0.00044871614844533604, "loss": 2.9565, "theoretical_loss": 4.049673617035414, "tokens_seen": 368115712 }, { "epoch": 0.11, "learning_rate": 0.0004487061183550652, "loss": 2.9942, "theoretical_loss": 4.0495918358857494, "tokens_seen": 368181248 }, { "epoch": 0.11, "learning_rate": 0.0004486960882647944, "loss": 3.1321, "theoretical_loss": 4.049510073366903, "tokens_seen": 368246784 }, { "epoch": 0.11, "learning_rate": 0.0004486860581745236, "loss": 3.0015, "theoretical_loss": 4.04942832947132, "tokens_seen": 368312320 }, { "epoch": 0.11, "learning_rate": 0.00044867602808425276, "loss": 2.8571, "theoretical_loss": 4.049346604191443, "tokens_seen": 368377856 }, { "epoch": 0.11, "learning_rate": 0.00044866599799398194, "loss": 2.8144, "theoretical_loss": 4.049264897519723, "tokens_seen": 368443392 }, { "epoch": 0.11, "learning_rate": 0.0004486559679037112, "loss": 3.0359, "theoretical_loss": 4.049183209448615, "tokens_seen": 368508928 }, { "epoch": 0.11, "learning_rate": 0.0004486459378134403, "loss": 3.0111, "theoretical_loss": 4.049101539970576, "tokens_seen": 368574464 }, { "epoch": 0.11, "objective/train/docs_used": 220098, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9974594116210938, "objective/train/theoretical_loss": 4.04901988907807, "objective/train/tokens_used": 389100000, "theoretical_loss": 4.04901988907807, "tokens_seen": 368640000 }, { "epoch": 0.11, "learning_rate": 0.00044863590772316954, "loss": 2.9455, "theoretical_loss": 4.04901988907807, "tokens_seen": 368640000 }, { "epoch": 0.11, "learning_rate": 0.00044862587763289867, "loss": 3.1135, "theoretical_loss": 4.048938256763565, "tokens_seen": 368705536 }, { "epoch": 0.11, "learning_rate": 0.0004486158475426279, "loss": 2.8557, "theoretical_loss": 4.048856643019532, "tokens_seen": 368771072 }, { "epoch": 0.11, "learning_rate": 0.0004486058174523571, "loss": 2.893, "theoretical_loss": 4.048775047838447, "tokens_seen": 368836608 }, { "epoch": 0.11, "learning_rate": 0.00044859578736208627, "loss": 3.0628, "theoretical_loss": 4.048693471212792, "tokens_seen": 368902144 }, { "epoch": 0.11, "learning_rate": 0.00044858575727181545, "loss": 2.9959, "theoretical_loss": 4.048611913135052, "tokens_seen": 368967680 }, { "epoch": 0.11, "learning_rate": 0.0004485757271815447, "loss": 3.0398, "theoretical_loss": 4.048530373597716, "tokens_seen": 369033216 }, { "epoch": 0.11, "learning_rate": 0.0004485656970912738, "loss": 3.0683, "theoretical_loss": 4.048448852593277, "tokens_seen": 369098752 }, { "epoch": 0.11, "learning_rate": 0.00044855566700100304, "loss": 2.8019, "theoretical_loss": 4.048367350114235, "tokens_seen": 369164288 }, { "epoch": 0.11, "learning_rate": 0.00044854563691073217, "loss": 3.3078, "theoretical_loss": 4.0482858661530905, "tokens_seen": 369229824 }, { "epoch": 0.11, "learning_rate": 0.0004485356068204614, "loss": 2.8911, "theoretical_loss": 4.048204400702351, "tokens_seen": 369295360 }, { "epoch": 0.11, "learning_rate": 0.0004485255767301906, "loss": 2.9811, "theoretical_loss": 4.04812295375453, "tokens_seen": 369360896 }, { "epoch": 0.11, "learning_rate": 0.00044851554663991977, "loss": 3.0611, "theoretical_loss": 4.04804152530214, "tokens_seen": 369426432 }, { "epoch": 0.11, "learning_rate": 0.00044850551654964895, "loss": 2.666, "theoretical_loss": 4.047960115337702, "tokens_seen": 369491968 }, { "epoch": 0.11, "learning_rate": 0.00044849548645937813, "loss": 2.9621, "theoretical_loss": 4.047878723853741, "tokens_seen": 369557504 }, { "epoch": 0.11, "learning_rate": 0.0004484854563691073, "loss": 3.1056, "theoretical_loss": 4.047797350842786, "tokens_seen": 369623040 }, { "epoch": 0.11, "learning_rate": 0.00044847542627883655, "loss": 2.9818, "theoretical_loss": 4.047715996297369, "tokens_seen": 369688576 }, { "epoch": 0.11, "learning_rate": 0.0004484653961885657, "loss": 3.1221, "theoretical_loss": 4.0476346602100275, "tokens_seen": 369754112 }, { "epoch": 0.11, "learning_rate": 0.0004484553660982949, "loss": 3.1221, "theoretical_loss": 4.047553342573304, "tokens_seen": 369819648 }, { "epoch": 0.11, "learning_rate": 0.00044844533600802404, "loss": 2.7862, "theoretical_loss": 4.047472043379743, "tokens_seen": 369885184 }, { "epoch": 0.11, "learning_rate": 0.00044843530591775327, "loss": 2.8429, "theoretical_loss": 4.047390762621897, "tokens_seen": 369950720 }, { "epoch": 0.11, "learning_rate": 0.00044842527582748245, "loss": 3.0292, "theoretical_loss": 4.04730950029232, "tokens_seen": 370016256 }, { "epoch": 0.11, "learning_rate": 0.00044841524573721163, "loss": 2.6929, "theoretical_loss": 4.047228256383569, "tokens_seen": 370081792 }, { "epoch": 0.11, "learning_rate": 0.0004484052156469408, "loss": 3.1469, "theoretical_loss": 4.04714703088821, "tokens_seen": 370147328 }, { "epoch": 0.11, "learning_rate": 0.00044839518555667005, "loss": 3.1589, "theoretical_loss": 4.0470658237988095, "tokens_seen": 370212864 }, { "epoch": 0.11, "objective/train/docs_used": 221122, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.944498300552368, "objective/train/theoretical_loss": 4.04698463510794, "objective/train/tokens_used": 390738400, "theoretical_loss": 4.04698463510794, "tokens_seen": 370278400 }, { "epoch": 0.11, "learning_rate": 0.0004483851554663992, "loss": 2.9615, "theoretical_loss": 4.04698463510794, "tokens_seen": 370278400 }, { "epoch": 0.11, "learning_rate": 0.0004483751253761284, "loss": 2.8592, "theoretical_loss": 4.046903464808177, "tokens_seen": 370343936 }, { "epoch": 0.11, "learning_rate": 0.00044836509528585754, "loss": 2.9014, "theoretical_loss": 4.046822312892102, "tokens_seen": 370409472 }, { "epoch": 0.11, "learning_rate": 0.0004483550651955868, "loss": 2.9322, "theoretical_loss": 4.046741179352299, "tokens_seen": 370475008 }, { "epoch": 0.11, "learning_rate": 0.00044834503510531596, "loss": 2.9574, "theoretical_loss": 4.046660064181359, "tokens_seen": 370540544 }, { "epoch": 0.11, "learning_rate": 0.00044833500501504514, "loss": 3.1144, "theoretical_loss": 4.0465789673718735, "tokens_seen": 370606080 }, { "epoch": 0.11, "learning_rate": 0.0004483249749247743, "loss": 3.0474, "theoretical_loss": 4.0464978889164405, "tokens_seen": 370671616 }, { "epoch": 0.11, "learning_rate": 0.0004483149448345035, "loss": 2.9841, "theoretical_loss": 4.046416828807663, "tokens_seen": 370737152 }, { "epoch": 0.11, "learning_rate": 0.0004483049147442327, "loss": 2.9314, "theoretical_loss": 4.046335787038148, "tokens_seen": 370802688 }, { "epoch": 0.11, "learning_rate": 0.0004482948846539619, "loss": 3.1382, "theoretical_loss": 4.046254763600503, "tokens_seen": 370868224 }, { "epoch": 0.11, "learning_rate": 0.00044828485456369104, "loss": 2.9275, "theoretical_loss": 4.046173758487347, "tokens_seen": 370933760 }, { "epoch": 0.11, "learning_rate": 0.0004482748244734203, "loss": 3.0695, "theoretical_loss": 4.046092771691297, "tokens_seen": 370999296 }, { "epoch": 0.11, "learning_rate": 0.0004482647943831494, "loss": 3.0102, "theoretical_loss": 4.046011803204976, "tokens_seen": 371064832 }, { "epoch": 0.11, "learning_rate": 0.00044825476429287864, "loss": 2.9351, "theoretical_loss": 4.045930853021014, "tokens_seen": 371130368 }, { "epoch": 0.11, "learning_rate": 0.0004482447342026078, "loss": 3.0805, "theoretical_loss": 4.045849921132042, "tokens_seen": 371195904 }, { "epoch": 0.11, "learning_rate": 0.000448234704112337, "loss": 2.8759, "theoretical_loss": 4.045769007530695, "tokens_seen": 371261440 }, { "epoch": 0.11, "learning_rate": 0.0004482246740220662, "loss": 3.2063, "theoretical_loss": 4.045688112209615, "tokens_seen": 371326976 }, { "epoch": 0.11, "learning_rate": 0.0004482146439317954, "loss": 2.9021, "theoretical_loss": 4.045607235161447, "tokens_seen": 371392512 }, { "epoch": 0.11, "learning_rate": 0.00044820461384152455, "loss": 2.8611, "theoretical_loss": 4.04552637637884, "tokens_seen": 371458048 }, { "epoch": 0.11, "learning_rate": 0.0004481945837512538, "loss": 3.1244, "theoretical_loss": 4.045445535854446, "tokens_seen": 371523584 }, { "epoch": 0.11, "learning_rate": 0.00044818455366098296, "loss": 2.9103, "theoretical_loss": 4.045364713580923, "tokens_seen": 371589120 }, { "epoch": 0.11, "learning_rate": 0.00044817452357071214, "loss": 2.9705, "theoretical_loss": 4.045283909550935, "tokens_seen": 371654656 }, { "epoch": 0.11, "learning_rate": 0.0004481644934804414, "loss": 3.0269, "theoretical_loss": 4.045203123757146, "tokens_seen": 371720192 }, { "epoch": 0.11, "learning_rate": 0.0004481544633901705, "loss": 3.0925, "theoretical_loss": 4.045122356192227, "tokens_seen": 371785728 }, { "epoch": 0.11, "learning_rate": 0.00044814443329989974, "loss": 3.0519, "theoretical_loss": 4.0450416068488515, "tokens_seen": 371851264 }, { "epoch": 0.11, "objective/train/docs_used": 221664, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7094762325286865, "objective/train/theoretical_loss": 4.044960875719701, "objective/train/tokens_used": 392376800, "theoretical_loss": 4.044960875719701, "tokens_seen": 371916800 }, { "epoch": 0.11, "learning_rate": 0.00044813440320962887, "loss": 2.7159, "theoretical_loss": 4.044960875719701, "tokens_seen": 371916800 }, { "epoch": 0.11, "learning_rate": 0.0004481243731193581, "loss": 2.9937, "theoretical_loss": 4.0448801627974555, "tokens_seen": 371982336 }, { "epoch": 0.11, "learning_rate": 0.0004481143430290873, "loss": 3.0136, "theoretical_loss": 4.044799468074804, "tokens_seen": 372047872 }, { "epoch": 0.11, "learning_rate": 0.00044810431293881647, "loss": 3.2027, "theoretical_loss": 4.044718791544438, "tokens_seen": 372113408 }, { "epoch": 0.11, "learning_rate": 0.00044809428284854565, "loss": 3.171, "theoretical_loss": 4.044638133199052, "tokens_seen": 372178944 }, { "epoch": 0.11, "learning_rate": 0.0004480842527582749, "loss": 2.9597, "theoretical_loss": 4.044557493031347, "tokens_seen": 372244480 }, { "epoch": 0.11, "learning_rate": 0.000448074222668004, "loss": 2.9182, "theoretical_loss": 4.044476871034027, "tokens_seen": 372310016 }, { "epoch": 0.11, "learning_rate": 0.00044806419257773324, "loss": 3.0117, "theoretical_loss": 4.0443962671998, "tokens_seen": 372375552 }, { "epoch": 0.11, "learning_rate": 0.00044805416248746237, "loss": 2.7842, "theoretical_loss": 4.044315681521379, "tokens_seen": 372441088 }, { "epoch": 0.11, "learning_rate": 0.0004480441323971916, "loss": 2.9061, "theoretical_loss": 4.04423511399148, "tokens_seen": 372506624 }, { "epoch": 0.11, "learning_rate": 0.0004480341023069208, "loss": 2.7469, "theoretical_loss": 4.044154564602826, "tokens_seen": 372572160 }, { "epoch": 0.11, "learning_rate": 0.00044802407221664997, "loss": 2.9115, "theoretical_loss": 4.044074033348139, "tokens_seen": 372637696 }, { "epoch": 0.11, "learning_rate": 0.00044801404212637915, "loss": 3.0106, "theoretical_loss": 4.043993520220151, "tokens_seen": 372703232 }, { "epoch": 0.11, "learning_rate": 0.00044800401203610833, "loss": 3.2011, "theoretical_loss": 4.0439130252115945, "tokens_seen": 372768768 }, { "epoch": 0.11, "learning_rate": 0.0004479939819458375, "loss": 3.1417, "theoretical_loss": 4.043832548315207, "tokens_seen": 372834304 }, { "epoch": 0.11, "learning_rate": 0.00044798395185556675, "loss": 3.0064, "theoretical_loss": 4.043752089523732, "tokens_seen": 372899840 }, { "epoch": 0.11, "learning_rate": 0.0004479739217652959, "loss": 3.0273, "theoretical_loss": 4.043671648829913, "tokens_seen": 372965376 }, { "epoch": 0.11, "learning_rate": 0.0004479638916750251, "loss": 2.785, "theoretical_loss": 4.043591226226503, "tokens_seen": 373030912 }, { "epoch": 0.11, "learning_rate": 0.00044795386158475424, "loss": 2.958, "theoretical_loss": 4.0435108217062545, "tokens_seen": 373096448 }, { "epoch": 0.11, "learning_rate": 0.00044794383149448347, "loss": 3.1277, "theoretical_loss": 4.043430435261927, "tokens_seen": 373161984 }, { "epoch": 0.11, "learning_rate": 0.00044793380140421265, "loss": 3.2189, "theoretical_loss": 4.043350066886283, "tokens_seen": 373227520 }, { "epoch": 0.11, "learning_rate": 0.00044792377131394183, "loss": 2.8955, "theoretical_loss": 4.043269716572091, "tokens_seen": 373293056 }, { "epoch": 0.11, "learning_rate": 0.000447913741223671, "loss": 2.713, "theoretical_loss": 4.04318938431212, "tokens_seen": 373358592 }, { "epoch": 0.11, "learning_rate": 0.00044790371113340025, "loss": 3.1045, "theoretical_loss": 4.043109070099146, "tokens_seen": 373424128 }, { "epoch": 0.11, "learning_rate": 0.0004478936810431294, "loss": 3.053, "theoretical_loss": 4.043028773925949, "tokens_seen": 373489664 }, { "epoch": 0.11, "objective/train/docs_used": 222906, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8282113075256348, "objective/train/theoretical_loss": 4.042948495785312, "objective/train/tokens_used": 394015200, "theoretical_loss": 4.042948495785312, "tokens_seen": 373555200 }, { "epoch": 0.11, "learning_rate": 0.0004478836509528586, "loss": 2.8803, "theoretical_loss": 4.042948495785312, "tokens_seen": 373555200 }, { "epoch": 0.11, "learning_rate": 0.00044787362086258774, "loss": 3.0613, "theoretical_loss": 4.042868235670023, "tokens_seen": 373620736 }, { "epoch": 0.11, "learning_rate": 0.000447863590772317, "loss": 2.8243, "theoretical_loss": 4.042787993572875, "tokens_seen": 373686272 }, { "epoch": 0.11, "learning_rate": 0.00044785356068204616, "loss": 3.0637, "theoretical_loss": 4.042707769486663, "tokens_seen": 373751808 }, { "epoch": 0.11, "learning_rate": 0.00044784353059177534, "loss": 3.0711, "theoretical_loss": 4.042627563404187, "tokens_seen": 373817344 }, { "epoch": 0.11, "learning_rate": 0.0004478335005015045, "loss": 2.85, "theoretical_loss": 4.042547375318251, "tokens_seen": 373882880 }, { "epoch": 0.11, "learning_rate": 0.0004478234704112337, "loss": 2.9246, "theoretical_loss": 4.042467205221665, "tokens_seen": 373948416 }, { "epoch": 0.11, "learning_rate": 0.0004478134403209629, "loss": 3.1001, "theoretical_loss": 4.042387053107241, "tokens_seen": 374013952 }, { "epoch": 0.11, "learning_rate": 0.0004478034102306921, "loss": 2.8072, "theoretical_loss": 4.042306918967795, "tokens_seen": 374079488 }, { "epoch": 0.11, "learning_rate": 0.00044779338014042124, "loss": 2.8558, "theoretical_loss": 4.042226802796149, "tokens_seen": 374145024 }, { "epoch": 0.11, "learning_rate": 0.0004477833500501505, "loss": 2.8778, "theoretical_loss": 4.042146704585127, "tokens_seen": 374210560 }, { "epoch": 0.11, "learning_rate": 0.0004477733199598796, "loss": 2.969, "theoretical_loss": 4.042066624327559, "tokens_seen": 374276096 }, { "epoch": 0.11, "learning_rate": 0.00044776328986960884, "loss": 3.0689, "theoretical_loss": 4.041986562016278, "tokens_seen": 374341632 }, { "epoch": 0.11, "learning_rate": 0.000447753259779338, "loss": 2.9921, "theoretical_loss": 4.041906517644121, "tokens_seen": 374407168 }, { "epoch": 0.11, "learning_rate": 0.0004477432296890672, "loss": 2.9593, "theoretical_loss": 4.0418264912039295, "tokens_seen": 374472704 }, { "epoch": 0.11, "learning_rate": 0.0004477331995987964, "loss": 2.8642, "theoretical_loss": 4.04174648268855, "tokens_seen": 374538240 }, { "epoch": 0.11, "learning_rate": 0.0004477231695085256, "loss": 2.8123, "theoretical_loss": 4.041666492090831, "tokens_seen": 374603776 }, { "epoch": 0.11, "learning_rate": 0.00044771313941825475, "loss": 2.7568, "theoretical_loss": 4.041586519403627, "tokens_seen": 374669312 }, { "epoch": 0.11, "learning_rate": 0.000447703109327984, "loss": 2.885, "theoretical_loss": 4.041506564619796, "tokens_seen": 374734848 }, { "epoch": 0.11, "learning_rate": 0.0004476930792377131, "loss": 2.7418, "theoretical_loss": 4.0414266277322, "tokens_seen": 374800384 }, { "epoch": 0.11, "learning_rate": 0.00044768304914744234, "loss": 3.0767, "theoretical_loss": 4.041346708733704, "tokens_seen": 374865920 }, { "epoch": 0.11, "learning_rate": 0.0004476730190571715, "loss": 3.0252, "theoretical_loss": 4.041266807617179, "tokens_seen": 374931456 }, { "epoch": 0.11, "learning_rate": 0.0004476629889669007, "loss": 2.6955, "theoretical_loss": 4.041186924375499, "tokens_seen": 374996992 }, { "epoch": 0.11, "learning_rate": 0.0004476529588766299, "loss": 2.6624, "theoretical_loss": 4.041107059001544, "tokens_seen": 375062528 }, { "epoch": 0.11, "learning_rate": 0.00044764292878635907, "loss": 3.0584, "theoretical_loss": 4.041027211488194, "tokens_seen": 375128064 }, { "epoch": 0.11, "objective/train/docs_used": 223164, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.815418004989624, "objective/train/theoretical_loss": 4.040947381828337, "objective/train/tokens_used": 395653600, "theoretical_loss": 4.040947381828337, "tokens_seen": 375193600 }, { "epoch": 0.11, "learning_rate": 0.00044763289869608825, "loss": 3.0237, "theoretical_loss": 4.040947381828337, "tokens_seen": 375193600 }, { "epoch": 0.11, "learning_rate": 0.0004476228686058175, "loss": 2.8633, "theoretical_loss": 4.040867570014862, "tokens_seen": 375259136 }, { "epoch": 0.11, "learning_rate": 0.0004476128385155466, "loss": 3.0381, "theoretical_loss": 4.040787776040666, "tokens_seen": 375324672 }, { "epoch": 0.11, "learning_rate": 0.00044760280842527585, "loss": 2.9233, "theoretical_loss": 4.040707999898647, "tokens_seen": 375390208 }, { "epoch": 0.11, "learning_rate": 0.00044759277833500503, "loss": 2.9864, "theoretical_loss": 4.040628241581706, "tokens_seen": 375455744 }, { "epoch": 0.11, "learning_rate": 0.0004475827482447342, "loss": 2.9743, "theoretical_loss": 4.0405485010827515, "tokens_seen": 375521280 }, { "epoch": 0.11, "learning_rate": 0.0004475727181544634, "loss": 3.1656, "theoretical_loss": 4.040468778394694, "tokens_seen": 375586816 }, { "epoch": 0.11, "learning_rate": 0.00044756268806419257, "loss": 2.7738, "theoretical_loss": 4.040389073510449, "tokens_seen": 375652352 }, { "epoch": 0.11, "learning_rate": 0.00044755265797392175, "loss": 3.043, "theoretical_loss": 4.040309386422934, "tokens_seen": 375717888 }, { "epoch": 0.11, "learning_rate": 0.000447542627883651, "loss": 3.0278, "theoretical_loss": 4.040229717125074, "tokens_seen": 375783424 }, { "epoch": 0.11, "learning_rate": 0.0004475325977933801, "loss": 3.1116, "theoretical_loss": 4.0401500656097955, "tokens_seen": 375848960 }, { "epoch": 0.11, "learning_rate": 0.00044752256770310935, "loss": 3.0037, "theoretical_loss": 4.040070431870029, "tokens_seen": 375914496 }, { "epoch": 0.11, "learning_rate": 0.0004475125376128385, "loss": 3.136, "theoretical_loss": 4.03999081589871, "tokens_seen": 375980032 }, { "epoch": 0.11, "learning_rate": 0.0004475025075225677, "loss": 3.1902, "theoretical_loss": 4.039911217688777, "tokens_seen": 376045568 }, { "epoch": 0.11, "learning_rate": 0.0004474924774322969, "loss": 2.9869, "theoretical_loss": 4.039831637233176, "tokens_seen": 376111104 }, { "epoch": 0.11, "learning_rate": 0.0004474824473420261, "loss": 2.8772, "theoretical_loss": 4.039752074524851, "tokens_seen": 376176640 }, { "epoch": 0.11, "learning_rate": 0.00044747241725175526, "loss": 2.94, "theoretical_loss": 4.039672529556755, "tokens_seen": 376242176 }, { "epoch": 0.11, "learning_rate": 0.00044746238716148444, "loss": 2.9516, "theoretical_loss": 4.039593002321844, "tokens_seen": 376307712 }, { "epoch": 0.11, "learning_rate": 0.0004474523570712136, "loss": 2.8799, "theoretical_loss": 4.039513492813076, "tokens_seen": 376373248 }, { "epoch": 0.11, "learning_rate": 0.00044744232698094285, "loss": 2.6616, "theoretical_loss": 4.039434001023415, "tokens_seen": 376438784 }, { "epoch": 0.11, "learning_rate": 0.00044743229689067204, "loss": 2.8112, "theoretical_loss": 4.039354526945829, "tokens_seen": 376504320 }, { "epoch": 0.11, "learning_rate": 0.0004474222668004012, "loss": 2.8584, "theoretical_loss": 4.039275070573288, "tokens_seen": 376569856 }, { "epoch": 0.11, "learning_rate": 0.00044741223671013045, "loss": 3.1688, "theoretical_loss": 4.039195631898768, "tokens_seen": 376635392 }, { "epoch": 0.11, "learning_rate": 0.0004474022066198596, "loss": 3.0963, "theoretical_loss": 4.03911621091525, "tokens_seen": 376700928 }, { "epoch": 0.11, "learning_rate": 0.0004473921765295888, "loss": 3.2786, "theoretical_loss": 4.039036807615716, "tokens_seen": 376766464 }, { "epoch": 0.11, "objective/train/docs_used": 224842, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6164164543151855, "objective/train/theoretical_loss": 4.038957421993153, "objective/train/tokens_used": 397292000, "theoretical_loss": 4.038957421993153, "tokens_seen": 376832000 }, { "epoch": 0.11, "learning_rate": 0.00044738214643931794, "loss": 3.0128, "theoretical_loss": 4.038957421993153, "tokens_seen": 376832000 }, { "epoch": 0.11, "learning_rate": 0.0004473721163490472, "loss": 2.5977, "theoretical_loss": 4.038878054040554, "tokens_seen": 376897536 }, { "epoch": 0.11, "learning_rate": 0.00044736208625877636, "loss": 3.1554, "theoretical_loss": 4.038798703750913, "tokens_seen": 376963072 }, { "epoch": 0.11, "learning_rate": 0.00044735205616850554, "loss": 2.8739, "theoretical_loss": 4.038719371117231, "tokens_seen": 377028608 }, { "epoch": 0.11, "learning_rate": 0.0004473420260782347, "loss": 2.8454, "theoretical_loss": 4.038640056132509, "tokens_seen": 377094144 }, { "epoch": 0.11, "learning_rate": 0.0004473319959879639, "loss": 3.0288, "theoretical_loss": 4.038560758789758, "tokens_seen": 377159680 }, { "epoch": 0.11, "learning_rate": 0.0004473219658976931, "loss": 3.0316, "theoretical_loss": 4.0384814790819865, "tokens_seen": 377225216 }, { "epoch": 0.11, "learning_rate": 0.0004473119358074223, "loss": 2.9039, "theoretical_loss": 4.038402217002211, "tokens_seen": 377290752 }, { "epoch": 0.11, "learning_rate": 0.00044730190571715144, "loss": 2.9524, "theoretical_loss": 4.038322972543451, "tokens_seen": 377356288 }, { "epoch": 0.11, "learning_rate": 0.0004472918756268807, "loss": 2.9518, "theoretical_loss": 4.03824374569873, "tokens_seen": 377421824 }, { "epoch": 0.11, "learning_rate": 0.0004472818455366098, "loss": 3.0085, "theoretical_loss": 4.038164536461075, "tokens_seen": 377487360 }, { "epoch": 0.11, "learning_rate": 0.00044727181544633904, "loss": 3.0309, "theoretical_loss": 4.038085344823516, "tokens_seen": 377552896 }, { "epoch": 0.11, "learning_rate": 0.0004472617853560682, "loss": 2.9822, "theoretical_loss": 4.038006170779091, "tokens_seen": 377618432 }, { "epoch": 0.11, "learning_rate": 0.0004472517552657974, "loss": 2.8212, "theoretical_loss": 4.037927014320838, "tokens_seen": 377683968 }, { "epoch": 0.11, "learning_rate": 0.0004472417251755266, "loss": 2.6607, "theoretical_loss": 4.037847875441798, "tokens_seen": 377749504 }, { "epoch": 0.11, "learning_rate": 0.0004472316950852558, "loss": 3.1658, "theoretical_loss": 4.037768754135023, "tokens_seen": 377815040 }, { "epoch": 0.11, "learning_rate": 0.00044722166499498495, "loss": 3.0641, "theoretical_loss": 4.037689650393559, "tokens_seen": 377880576 }, { "epoch": 0.11, "learning_rate": 0.0004472116349047142, "loss": 3.0602, "theoretical_loss": 4.037610564210464, "tokens_seen": 377946112 }, { "epoch": 0.11, "learning_rate": 0.0004472016048144433, "loss": 2.7781, "theoretical_loss": 4.037531495578797, "tokens_seen": 378011648 }, { "epoch": 0.11, "learning_rate": 0.00044719157472417254, "loss": 2.911, "theoretical_loss": 4.03745244449162, "tokens_seen": 378077184 }, { "epoch": 0.11, "learning_rate": 0.0004471815446339017, "loss": 3.1229, "theoretical_loss": 4.037373410942001, "tokens_seen": 378142720 }, { "epoch": 0.11, "learning_rate": 0.0004471715145436309, "loss": 2.8328, "theoretical_loss": 4.037294394923009, "tokens_seen": 378208256 }, { "epoch": 0.11, "learning_rate": 0.0004471614844533601, "loss": 3.0088, "theoretical_loss": 4.037215396427721, "tokens_seen": 378273792 }, { "epoch": 0.11, "learning_rate": 0.00044715145436308927, "loss": 3.2103, "theoretical_loss": 4.037136415449214, "tokens_seen": 378339328 }, { "epoch": 0.11, "learning_rate": 0.00044714142427281845, "loss": 2.897, "theoretical_loss": 4.037057451980571, "tokens_seen": 378404864 }, { "epoch": 0.11, "objective/train/docs_used": 225570, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.853151321411133, "objective/train/theoretical_loss": 4.036978506014879, "objective/train/tokens_used": 398930400, "theoretical_loss": 4.036978506014879, "tokens_seen": 378470400 }, { "epoch": 0.11, "learning_rate": 0.0004471313941825477, "loss": 3.0427, "theoretical_loss": 4.036978506014879, "tokens_seen": 378470400 }, { "epoch": 0.11, "learning_rate": 0.0004471213640922768, "loss": 2.9945, "theoretical_loss": 4.036899577545229, "tokens_seen": 378535936 }, { "epoch": 0.11, "learning_rate": 0.00044711133400200605, "loss": 3.0209, "theoretical_loss": 4.036820666564714, "tokens_seen": 378601472 }, { "epoch": 0.11, "learning_rate": 0.00044710130391173523, "loss": 2.7759, "theoretical_loss": 4.036741773066434, "tokens_seen": 378667008 }, { "epoch": 0.11, "learning_rate": 0.0004470912738214644, "loss": 3.096, "theoretical_loss": 4.03666289704349, "tokens_seen": 378732544 }, { "epoch": 0.11, "learning_rate": 0.0004470812437311936, "loss": 2.958, "theoretical_loss": 4.036584038488988, "tokens_seen": 378798080 }, { "epoch": 0.11, "learning_rate": 0.00044707121364092277, "loss": 2.899, "theoretical_loss": 4.03650519739604, "tokens_seen": 378863616 }, { "epoch": 0.11, "learning_rate": 0.00044706118355065195, "loss": 3.0908, "theoretical_loss": 4.036426373757759, "tokens_seen": 378929152 }, { "epoch": 0.11, "learning_rate": 0.0004470511534603812, "loss": 3.1033, "theoretical_loss": 4.036347567567262, "tokens_seen": 378994688 }, { "epoch": 0.11, "learning_rate": 0.0004470411233701103, "loss": 2.6781, "theoretical_loss": 4.0362687788176705, "tokens_seen": 379060224 }, { "epoch": 0.11, "learning_rate": 0.00044703109327983955, "loss": 3.2151, "theoretical_loss": 4.036190007502112, "tokens_seen": 379125760 }, { "epoch": 0.11, "learning_rate": 0.0004470210631895687, "loss": 2.9318, "theoretical_loss": 4.036111253613715, "tokens_seen": 379191296 }, { "epoch": 0.11, "learning_rate": 0.0004470110330992979, "loss": 2.9802, "theoretical_loss": 4.036032517145614, "tokens_seen": 379256832 }, { "epoch": 0.11, "learning_rate": 0.0004470010030090271, "loss": 2.8508, "theoretical_loss": 4.0359537980909455, "tokens_seen": 379322368 }, { "epoch": 0.11, "learning_rate": 0.0004469909729187563, "loss": 3.1785, "theoretical_loss": 4.03587509644285, "tokens_seen": 379387904 }, { "epoch": 0.11, "learning_rate": 0.00044698094282848546, "loss": 3.0789, "theoretical_loss": 4.0357964121944745, "tokens_seen": 379453440 }, { "epoch": 0.12, "learning_rate": 0.00044697091273821464, "loss": 2.9847, "theoretical_loss": 4.035717745338967, "tokens_seen": 379518976 }, { "epoch": 0.12, "learning_rate": 0.0004469608826479438, "loss": 2.8707, "theoretical_loss": 4.035639095869481, "tokens_seen": 379584512 }, { "epoch": 0.12, "learning_rate": 0.00044695085255767305, "loss": 3.1781, "theoretical_loss": 4.035560463779173, "tokens_seen": 379650048 }, { "epoch": 0.12, "learning_rate": 0.0004469408224674022, "loss": 3.2176, "theoretical_loss": 4.0354818490612026, "tokens_seen": 379715584 }, { "epoch": 0.12, "learning_rate": 0.0004469307923771314, "loss": 2.9979, "theoretical_loss": 4.035403251708736, "tokens_seen": 379781120 }, { "epoch": 0.12, "learning_rate": 0.0004469207622868606, "loss": 3.2267, "theoretical_loss": 4.035324671714942, "tokens_seen": 379846656 }, { "epoch": 0.12, "learning_rate": 0.0004469107321965898, "loss": 3.1496, "theoretical_loss": 4.0352461090729905, "tokens_seen": 379912192 }, { "epoch": 0.12, "learning_rate": 0.00044690070210631896, "loss": 2.9406, "theoretical_loss": 4.03516756377606, "tokens_seen": 379977728 }, { "epoch": 0.12, "learning_rate": 0.00044689067201604814, "loss": 2.9065, "theoretical_loss": 4.035089035817329, "tokens_seen": 380043264 }, { "epoch": 0.12, "objective/train/docs_used": 226798, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8520052433013916, "objective/train/theoretical_loss": 4.035010525189982, "objective/train/tokens_used": 400568800, "theoretical_loss": 4.035010525189982, "tokens_seen": 380108800 }, { "epoch": 0.12, "learning_rate": 0.0004468806419257773, "loss": 2.7012, "theoretical_loss": 4.035010525189982, "tokens_seen": 380108800 }, { "epoch": 0.12, "learning_rate": 0.00044687061183550656, "loss": 3.035, "theoretical_loss": 4.034932031887207, "tokens_seen": 380174336 }, { "epoch": 0.12, "learning_rate": 0.0004468605817452357, "loss": 2.9595, "theoretical_loss": 4.0348535559021945, "tokens_seen": 380239872 }, { "epoch": 0.12, "learning_rate": 0.0004468505516549649, "loss": 2.9639, "theoretical_loss": 4.034775097228142, "tokens_seen": 380305408 }, { "epoch": 0.12, "learning_rate": 0.00044684052156469405, "loss": 3.1619, "theoretical_loss": 4.034696655858246, "tokens_seen": 380370944 }, { "epoch": 0.12, "learning_rate": 0.0004468304914744233, "loss": 2.9525, "theoretical_loss": 4.0346182317857116, "tokens_seen": 380436480 }, { "epoch": 0.12, "learning_rate": 0.00044682046138415246, "loss": 3.0604, "theoretical_loss": 4.034539825003746, "tokens_seen": 380502016 }, { "epoch": 0.12, "learning_rate": 0.00044681043129388164, "loss": 2.8797, "theoretical_loss": 4.034461435505557, "tokens_seen": 380567552 }, { "epoch": 0.12, "learning_rate": 0.0004468004012036108, "loss": 2.8979, "theoretical_loss": 4.034383063284364, "tokens_seen": 380633088 }, { "epoch": 0.12, "learning_rate": 0.00044679037111334, "loss": 3.0024, "theoretical_loss": 4.034304708333382, "tokens_seen": 380698624 }, { "epoch": 0.12, "learning_rate": 0.0004467803410230692, "loss": 3.0039, "theoretical_loss": 4.0342263706458334, "tokens_seen": 380764160 }, { "epoch": 0.12, "learning_rate": 0.0004467703109327984, "loss": 2.9339, "theoretical_loss": 4.034148050214946, "tokens_seen": 380829696 }, { "epoch": 0.12, "learning_rate": 0.00044676028084252755, "loss": 3.0316, "theoretical_loss": 4.034069747033948, "tokens_seen": 380895232 }, { "epoch": 0.12, "learning_rate": 0.0004467502507522568, "loss": 3.1285, "theoretical_loss": 4.0339914610960745, "tokens_seen": 380960768 }, { "epoch": 0.12, "learning_rate": 0.00044674022066198597, "loss": 3.0281, "theoretical_loss": 4.033913192394563, "tokens_seen": 381026304 }, { "epoch": 0.12, "learning_rate": 0.00044673019057171515, "loss": 2.8929, "theoretical_loss": 4.033834940922653, "tokens_seen": 381091840 }, { "epoch": 0.12, "learning_rate": 0.00044672016048144433, "loss": 2.6042, "theoretical_loss": 4.033756706673593, "tokens_seen": 381157376 }, { "epoch": 0.12, "learning_rate": 0.0004467101303911735, "loss": 3.0573, "theoretical_loss": 4.033678489640629, "tokens_seen": 381222912 }, { "epoch": 0.12, "learning_rate": 0.0004467001003009027, "loss": 2.9445, "theoretical_loss": 4.033600289817015, "tokens_seen": 381288448 }, { "epoch": 0.12, "learning_rate": 0.0004466900702106319, "loss": 3.0191, "theoretical_loss": 4.033522107196007, "tokens_seen": 381353984 }, { "epoch": 0.12, "learning_rate": 0.0004466800401203611, "loss": 2.9899, "theoretical_loss": 4.033443941770867, "tokens_seen": 381419520 }, { "epoch": 0.12, "learning_rate": 0.0004466700100300903, "loss": 2.7601, "theoretical_loss": 4.033365793534857, "tokens_seen": 381485056 }, { "epoch": 0.12, "learning_rate": 0.00044665997993981947, "loss": 3.0316, "theoretical_loss": 4.033287662481247, "tokens_seen": 381550592 }, { "epoch": 0.12, "learning_rate": 0.00044664994984954865, "loss": 2.9178, "theoretical_loss": 4.033209548603308, "tokens_seen": 381616128 }, { "epoch": 0.12, "learning_rate": 0.0004466399197592779, "loss": 3.1146, "theoretical_loss": 4.033131451894317, "tokens_seen": 381681664 }, { "epoch": 0.12, "objective/train/docs_used": 227456, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.782240867614746, "objective/train/theoretical_loss": 4.03305337234755, "objective/train/tokens_used": 402207200, "theoretical_loss": 4.03305337234755, "tokens_seen": 381747200 }, { "epoch": 0.12, "learning_rate": 0.000446629889669007, "loss": 2.8048, "theoretical_loss": 4.03305337234755, "tokens_seen": 381747200 }, { "epoch": 0.12, "learning_rate": 0.00044661985957873625, "loss": 3.125, "theoretical_loss": 4.032975309956292, "tokens_seen": 381812736 }, { "epoch": 0.12, "learning_rate": 0.00044660982948846543, "loss": 3.0035, "theoretical_loss": 4.0328972647138315, "tokens_seen": 381878272 }, { "epoch": 0.12, "learning_rate": 0.0004465997993981946, "loss": 2.9588, "theoretical_loss": 4.032819236613457, "tokens_seen": 381943808 }, { "epoch": 0.12, "learning_rate": 0.0004465897693079238, "loss": 3.0603, "theoretical_loss": 4.032741225648463, "tokens_seen": 382009344 }, { "epoch": 0.12, "learning_rate": 0.00044657973921765297, "loss": 3.2, "theoretical_loss": 4.03266323181215, "tokens_seen": 382074880 }, { "epoch": 0.12, "learning_rate": 0.00044656970912738215, "loss": 2.8238, "theoretical_loss": 4.0325852550978185, "tokens_seen": 382140416 }, { "epoch": 0.12, "learning_rate": 0.0004465596790371114, "loss": 3.0385, "theoretical_loss": 4.032507295498774, "tokens_seen": 382205952 }, { "epoch": 0.12, "learning_rate": 0.0004465496489468405, "loss": 2.787, "theoretical_loss": 4.032429353008326, "tokens_seen": 382271488 }, { "epoch": 0.12, "learning_rate": 0.00044653961885656975, "loss": 3.1649, "theoretical_loss": 4.032351427619789, "tokens_seen": 382337024 }, { "epoch": 0.12, "learning_rate": 0.0004465295887662989, "loss": 3.0231, "theoretical_loss": 4.03227351932648, "tokens_seen": 382402560 }, { "epoch": 0.12, "learning_rate": 0.0004465195586760281, "loss": 3.1729, "theoretical_loss": 4.032195628121718, "tokens_seen": 382468096 }, { "epoch": 0.12, "learning_rate": 0.0004465095285857573, "loss": 2.9617, "theoretical_loss": 4.032117753998829, "tokens_seen": 382533632 }, { "epoch": 0.12, "learning_rate": 0.0004464994984954865, "loss": 2.8875, "theoretical_loss": 4.032039896951142, "tokens_seen": 382599168 }, { "epoch": 0.12, "learning_rate": 0.00044648946840521566, "loss": 2.9509, "theoretical_loss": 4.031962056971988, "tokens_seen": 382664704 }, { "epoch": 0.12, "learning_rate": 0.00044647943831494484, "loss": 2.7274, "theoretical_loss": 4.0318842340547025, "tokens_seen": 382730240 }, { "epoch": 0.12, "learning_rate": 0.000446469408224674, "loss": 2.8627, "theoretical_loss": 4.031806428192627, "tokens_seen": 382795776 }, { "epoch": 0.12, "learning_rate": 0.00044645937813440325, "loss": 2.9433, "theoretical_loss": 4.031728639379103, "tokens_seen": 382861312 }, { "epoch": 0.12, "learning_rate": 0.0004464493480441324, "loss": 3.0113, "theoretical_loss": 4.031650867607478, "tokens_seen": 382926848 }, { "epoch": 0.12, "learning_rate": 0.0004464393179538616, "loss": 3.0806, "theoretical_loss": 4.031573112871103, "tokens_seen": 382992384 }, { "epoch": 0.12, "learning_rate": 0.0004464292878635908, "loss": 2.9037, "theoretical_loss": 4.031495375163333, "tokens_seen": 383057920 }, { "epoch": 0.12, "learning_rate": 0.00044641925777332, "loss": 3.1862, "theoretical_loss": 4.031417654477526, "tokens_seen": 383123456 }, { "epoch": 0.12, "learning_rate": 0.00044640922768304916, "loss": 2.9205, "theoretical_loss": 4.031339950807043, "tokens_seen": 383188992 }, { "epoch": 0.12, "learning_rate": 0.00044639919759277834, "loss": 2.8033, "theoretical_loss": 4.03126226414525, "tokens_seen": 383254528 }, { "epoch": 0.12, "learning_rate": 0.0004463891675025075, "loss": 2.9103, "theoretical_loss": 4.031184594485517, "tokens_seen": 383320064 }, { "epoch": 0.12, "objective/train/docs_used": 228876, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.048848867416382, "objective/train/theoretical_loss": 4.031106941821218, "objective/train/tokens_used": 403845600, "theoretical_loss": 4.031106941821218, "tokens_seen": 383385600 }, { "epoch": 0.12, "learning_rate": 0.00044637913741223676, "loss": 2.8579, "theoretical_loss": 4.031106941821218, "tokens_seen": 383385600 }, { "epoch": 0.12, "learning_rate": 0.0004463691073219659, "loss": 3.0792, "theoretical_loss": 4.031029306145728, "tokens_seen": 383451136 }, { "epoch": 0.12, "learning_rate": 0.0004463590772316951, "loss": 3.1489, "theoretical_loss": 4.030951687452429, "tokens_seen": 383516672 }, { "epoch": 0.12, "learning_rate": 0.00044634904714142425, "loss": 3.1007, "theoretical_loss": 4.0308740857347045, "tokens_seen": 383582208 }, { "epoch": 0.12, "learning_rate": 0.0004463390170511535, "loss": 3.1964, "theoretical_loss": 4.030796500985943, "tokens_seen": 383647744 }, { "epoch": 0.12, "learning_rate": 0.00044632898696088266, "loss": 3.0761, "theoretical_loss": 4.030718933199536, "tokens_seen": 383713280 }, { "epoch": 0.12, "learning_rate": 0.00044631895687061184, "loss": 2.9627, "theoretical_loss": 4.030641382368878, "tokens_seen": 383778816 }, { "epoch": 0.12, "learning_rate": 0.000446308926780341, "loss": 2.9387, "theoretical_loss": 4.03056384848737, "tokens_seen": 383844352 }, { "epoch": 0.12, "learning_rate": 0.0004462988966900702, "loss": 2.7182, "theoretical_loss": 4.030486331548413, "tokens_seen": 383909888 }, { "epoch": 0.12, "learning_rate": 0.0004462888665997994, "loss": 3.1237, "theoretical_loss": 4.030408831545414, "tokens_seen": 383975424 }, { "epoch": 0.12, "learning_rate": 0.0004462788365095286, "loss": 2.9458, "theoretical_loss": 4.030331348471784, "tokens_seen": 384040960 }, { "epoch": 0.12, "learning_rate": 0.00044626880641925775, "loss": 3.0304, "theoretical_loss": 4.030253882320936, "tokens_seen": 384106496 }, { "epoch": 0.12, "learning_rate": 0.000446258776328987, "loss": 3.0564, "theoretical_loss": 4.030176433086289, "tokens_seen": 384172032 }, { "epoch": 0.12, "learning_rate": 0.00044624874623871617, "loss": 3.1166, "theoretical_loss": 4.030099000761261, "tokens_seen": 384237568 }, { "epoch": 0.12, "learning_rate": 0.00044623871614844535, "loss": 3.0936, "theoretical_loss": 4.030021585339281, "tokens_seen": 384303104 }, { "epoch": 0.12, "learning_rate": 0.00044622868605817453, "loss": 2.9371, "theoretical_loss": 4.029944186813774, "tokens_seen": 384368640 }, { "epoch": 0.12, "learning_rate": 0.0004462186559679037, "loss": 3.0269, "theoretical_loss": 4.029866805178175, "tokens_seen": 384434176 }, { "epoch": 0.12, "learning_rate": 0.0004462086258776329, "loss": 2.9255, "theoretical_loss": 4.029789440425917, "tokens_seen": 384499712 }, { "epoch": 0.12, "learning_rate": 0.0004461985957873621, "loss": 3.0704, "theoretical_loss": 4.029712092550443, "tokens_seen": 384565248 }, { "epoch": 0.12, "learning_rate": 0.00044618856569709125, "loss": 2.9137, "theoretical_loss": 4.029634761545195, "tokens_seen": 384630784 }, { "epoch": 0.12, "learning_rate": 0.0004461785356068205, "loss": 3.0362, "theoretical_loss": 4.029557447403618, "tokens_seen": 384696320 }, { "epoch": 0.12, "learning_rate": 0.0004461685055165496, "loss": 3.0371, "theoretical_loss": 4.029480150119165, "tokens_seen": 384761856 }, { "epoch": 0.12, "learning_rate": 0.00044615847542627885, "loss": 2.7249, "theoretical_loss": 4.029402869685288, "tokens_seen": 384827392 }, { "epoch": 0.12, "learning_rate": 0.00044614844533600803, "loss": 3.0639, "theoretical_loss": 4.0293256060954485, "tokens_seen": 384892928 }, { "epoch": 0.12, "learning_rate": 0.0004461384152457372, "loss": 2.8059, "theoretical_loss": 4.029248359343105, "tokens_seen": 384958464 }, { "epoch": 0.12, "objective/train/docs_used": 229454, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6626176834106445, "objective/train/theoretical_loss": 4.029171129421724, "objective/train/tokens_used": 405484000, "theoretical_loss": 4.029171129421724, "tokens_seen": 385024000 }, { "epoch": 0.12, "learning_rate": 0.0004461283851554664, "loss": 3.0008, "theoretical_loss": 4.029171129421724, "tokens_seen": 385024000 }, { "epoch": 0.12, "learning_rate": 0.00044611835506519563, "loss": 2.9454, "theoretical_loss": 4.029093916324774, "tokens_seen": 385089536 }, { "epoch": 0.12, "learning_rate": 0.00044610832497492476, "loss": 2.9709, "theoretical_loss": 4.0290167200457265, "tokens_seen": 385155072 }, { "epoch": 0.12, "learning_rate": 0.000446098294884654, "loss": 2.8698, "theoretical_loss": 4.02893954057806, "tokens_seen": 385220608 }, { "epoch": 0.12, "learning_rate": 0.0004460882647943831, "loss": 3.0199, "theoretical_loss": 4.028862377915252, "tokens_seen": 385286144 }, { "epoch": 0.12, "learning_rate": 0.00044607823470411235, "loss": 3.2099, "theoretical_loss": 4.0287852320507875, "tokens_seen": 385351680 }, { "epoch": 0.12, "learning_rate": 0.00044606820461384153, "loss": 3.0059, "theoretical_loss": 4.028708102978152, "tokens_seen": 385417216 }, { "epoch": 0.12, "learning_rate": 0.0004460581745235707, "loss": 2.9014, "theoretical_loss": 4.028630990690838, "tokens_seen": 385482752 }, { "epoch": 0.12, "learning_rate": 0.0004460481444332999, "loss": 2.9839, "theoretical_loss": 4.028553895182339, "tokens_seen": 385548288 }, { "epoch": 0.12, "learning_rate": 0.0004460381143430291, "loss": 2.9966, "theoretical_loss": 4.028476816446153, "tokens_seen": 385613824 }, { "epoch": 0.12, "learning_rate": 0.00044602808425275826, "loss": 2.9376, "theoretical_loss": 4.028399754475781, "tokens_seen": 385679360 }, { "epoch": 0.12, "learning_rate": 0.0004460180541624875, "loss": 2.8859, "theoretical_loss": 4.028322709264729, "tokens_seen": 385744896 }, { "epoch": 0.12, "learning_rate": 0.0004460080240722166, "loss": 2.9689, "theoretical_loss": 4.0282456808065055, "tokens_seen": 385810432 }, { "epoch": 0.12, "learning_rate": 0.00044599799398194586, "loss": 3.1434, "theoretical_loss": 4.028168669094623, "tokens_seen": 385875968 }, { "epoch": 0.12, "learning_rate": 0.000445987963891675, "loss": 3.0785, "theoretical_loss": 4.028091674122598, "tokens_seen": 385941504 }, { "epoch": 0.12, "learning_rate": 0.0004459779338014042, "loss": 2.9047, "theoretical_loss": 4.0280146958839484, "tokens_seen": 386007040 }, { "epoch": 0.12, "learning_rate": 0.0004459679037111334, "loss": 3.2321, "theoretical_loss": 4.0279377343722, "tokens_seen": 386072576 }, { "epoch": 0.12, "learning_rate": 0.0004459578736208626, "loss": 2.9964, "theoretical_loss": 4.027860789580878, "tokens_seen": 386138112 }, { "epoch": 0.12, "learning_rate": 0.00044594784353059176, "loss": 2.7246, "theoretical_loss": 4.027783861503513, "tokens_seen": 386203648 }, { "epoch": 0.12, "learning_rate": 0.000445937813440321, "loss": 2.9984, "theoretical_loss": 4.0277069501336396, "tokens_seen": 386269184 }, { "epoch": 0.12, "learning_rate": 0.0004459277833500502, "loss": 2.9427, "theoretical_loss": 4.027630055464795, "tokens_seen": 386334720 }, { "epoch": 0.12, "learning_rate": 0.00044591775325977936, "loss": 2.9795, "theoretical_loss": 4.027553177490521, "tokens_seen": 386400256 }, { "epoch": 0.12, "learning_rate": 0.00044590772316950854, "loss": 3.1006, "theoretical_loss": 4.027476316204362, "tokens_seen": 386465792 }, { "epoch": 0.12, "learning_rate": 0.0004458976930792377, "loss": 3.0388, "theoretical_loss": 4.027399471599867, "tokens_seen": 386531328 }, { "epoch": 0.12, "learning_rate": 0.00044588766298896696, "loss": 3.1798, "theoretical_loss": 4.027322643670587, "tokens_seen": 386596864 }, { "epoch": 0.12, "objective/train/docs_used": 230121, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1686201095581055, "objective/train/theoretical_loss": 4.027245832410079, "objective/train/tokens_used": 407122400, "theoretical_loss": 4.027245832410079, "tokens_seen": 386662400 }, { "epoch": 0.12, "learning_rate": 0.0004458776328986961, "loss": 2.9233, "theoretical_loss": 4.027245832410079, "tokens_seen": 386662400 }, { "epoch": 0.12, "learning_rate": 0.0004458676028084253, "loss": 2.8835, "theoretical_loss": 4.027169037811901, "tokens_seen": 386727936 }, { "epoch": 0.12, "learning_rate": 0.00044585757271815445, "loss": 2.8742, "theoretical_loss": 4.027092259869616, "tokens_seen": 386793472 }, { "epoch": 0.12, "learning_rate": 0.0004458475426278837, "loss": 2.8106, "theoretical_loss": 4.027015498576791, "tokens_seen": 386859008 }, { "epoch": 0.12, "learning_rate": 0.00044583751253761286, "loss": 2.8313, "theoretical_loss": 4.026938753926997, "tokens_seen": 386924544 }, { "epoch": 0.12, "learning_rate": 0.00044582748244734204, "loss": 2.9577, "theoretical_loss": 4.026862025913804, "tokens_seen": 386990080 }, { "epoch": 0.12, "learning_rate": 0.0004458174523570712, "loss": 3.0805, "theoretical_loss": 4.026785314530792, "tokens_seen": 387055616 }, { "epoch": 0.12, "learning_rate": 0.0004458074222668004, "loss": 3.0757, "theoretical_loss": 4.026708619771541, "tokens_seen": 387121152 }, { "epoch": 0.12, "learning_rate": 0.0004457973921765296, "loss": 2.9777, "theoretical_loss": 4.026631941629635, "tokens_seen": 387186688 }, { "epoch": 0.12, "learning_rate": 0.0004457873620862588, "loss": 3.1366, "theoretical_loss": 4.026555280098661, "tokens_seen": 387252224 }, { "epoch": 0.12, "learning_rate": 0.00044577733199598795, "loss": 3.09, "theoretical_loss": 4.026478635172212, "tokens_seen": 387317760 }, { "epoch": 0.12, "learning_rate": 0.0004457673019057172, "loss": 2.9166, "theoretical_loss": 4.026402006843882, "tokens_seen": 387383296 }, { "epoch": 0.12, "learning_rate": 0.00044575727181544637, "loss": 2.8882, "theoretical_loss": 4.0263253951072695, "tokens_seen": 387448832 }, { "epoch": 0.12, "learning_rate": 0.00044574724172517555, "loss": 2.7889, "theoretical_loss": 4.026248799955976, "tokens_seen": 387514368 }, { "epoch": 0.12, "learning_rate": 0.00044573721163490473, "loss": 2.8333, "theoretical_loss": 4.026172221383607, "tokens_seen": 387579904 }, { "epoch": 0.12, "learning_rate": 0.0004457271815446339, "loss": 2.7393, "theoretical_loss": 4.026095659383773, "tokens_seen": 387645440 }, { "epoch": 0.12, "learning_rate": 0.0004457171514543631, "loss": 2.9765, "theoretical_loss": 4.026019113950086, "tokens_seen": 387710976 }, { "epoch": 0.12, "learning_rate": 0.0004457071213640923, "loss": 2.9468, "theoretical_loss": 4.025942585076161, "tokens_seen": 387776512 }, { "epoch": 0.12, "learning_rate": 0.00044569709127382145, "loss": 2.9553, "theoretical_loss": 4.025866072755619, "tokens_seen": 387842048 }, { "epoch": 0.12, "learning_rate": 0.0004456870611835507, "loss": 3.0161, "theoretical_loss": 4.025789576982083, "tokens_seen": 387907584 }, { "epoch": 0.12, "learning_rate": 0.0004456770310932798, "loss": 2.9051, "theoretical_loss": 4.02571309774918, "tokens_seen": 387973120 }, { "epoch": 0.12, "learning_rate": 0.00044566700100300905, "loss": 2.9076, "theoretical_loss": 4.025636635050539, "tokens_seen": 388038656 }, { "epoch": 0.12, "learning_rate": 0.00044565697091273823, "loss": 3.0276, "theoretical_loss": 4.025560188879796, "tokens_seen": 388104192 }, { "epoch": 0.12, "learning_rate": 0.0004456469408224674, "loss": 2.911, "theoretical_loss": 4.025483759230585, "tokens_seen": 388169728 }, { "epoch": 0.12, "learning_rate": 0.0004456369107321966, "loss": 2.8527, "theoretical_loss": 4.025407346096551, "tokens_seen": 388235264 }, { "epoch": 0.12, "objective/train/docs_used": 231531, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4107649326324463, "objective/train/theoretical_loss": 4.025330949471336, "objective/train/tokens_used": 408760800, "theoretical_loss": 4.025330949471336, "tokens_seen": 388300800 }, { "epoch": 0.12, "learning_rate": 0.00044562688064192583, "loss": 2.909, "theoretical_loss": 4.025330949471336, "tokens_seen": 388300800 }, { "epoch": 0.12, "learning_rate": 0.00044561685055165496, "loss": 3.0747, "theoretical_loss": 4.025254569348588, "tokens_seen": 388366336 }, { "epoch": 0.12, "learning_rate": 0.0004456068204613842, "loss": 3.0523, "theoretical_loss": 4.025178205721959, "tokens_seen": 388431872 }, { "epoch": 0.12, "learning_rate": 0.0004455967903711133, "loss": 2.9877, "theoretical_loss": 4.025101858585104, "tokens_seen": 388497408 }, { "epoch": 0.12, "learning_rate": 0.00044558676028084255, "loss": 2.9718, "theoretical_loss": 4.025025527931681, "tokens_seen": 388562944 }, { "epoch": 0.12, "learning_rate": 0.00044557673019057173, "loss": 3.0657, "theoretical_loss": 4.0249492137553515, "tokens_seen": 388628480 }, { "epoch": 0.12, "learning_rate": 0.0004455667001003009, "loss": 2.8732, "theoretical_loss": 4.024872916049782, "tokens_seen": 388694016 }, { "epoch": 0.12, "learning_rate": 0.0004455566700100301, "loss": 3.1106, "theoretical_loss": 4.0247966348086415, "tokens_seen": 388759552 }, { "epoch": 0.12, "learning_rate": 0.0004455466399197593, "loss": 3.0527, "theoretical_loss": 4.0247203700256025, "tokens_seen": 388825088 }, { "epoch": 0.12, "learning_rate": 0.00044553660982948846, "loss": 3.0707, "theoretical_loss": 4.02464412169434, "tokens_seen": 388890624 }, { "epoch": 0.12, "learning_rate": 0.0004455265797392177, "loss": 2.846, "theoretical_loss": 4.024567889808534, "tokens_seen": 388956160 }, { "epoch": 0.12, "learning_rate": 0.0004455165496489468, "loss": 3.1694, "theoretical_loss": 4.024491674361868, "tokens_seen": 389021696 }, { "epoch": 0.12, "learning_rate": 0.00044550651955867606, "loss": 2.8077, "theoretical_loss": 4.024415475348027, "tokens_seen": 389087232 }, { "epoch": 0.12, "learning_rate": 0.0004454964894684052, "loss": 2.9521, "theoretical_loss": 4.024339292760703, "tokens_seen": 389152768 }, { "epoch": 0.12, "learning_rate": 0.0004454864593781344, "loss": 2.8418, "theoretical_loss": 4.024263126593588, "tokens_seen": 389218304 }, { "epoch": 0.12, "learning_rate": 0.0004454764292878636, "loss": 2.946, "theoretical_loss": 4.0241869768403795, "tokens_seen": 389283840 }, { "epoch": 0.12, "learning_rate": 0.0004454663991975928, "loss": 2.8275, "theoretical_loss": 4.024110843494777, "tokens_seen": 389349376 }, { "epoch": 0.12, "learning_rate": 0.00044545636910732196, "loss": 2.9606, "theoretical_loss": 4.024034726550486, "tokens_seen": 389414912 }, { "epoch": 0.12, "learning_rate": 0.0004454463390170512, "loss": 2.7693, "theoretical_loss": 4.023958626001212, "tokens_seen": 389480448 }, { "epoch": 0.12, "learning_rate": 0.0004454363089267803, "loss": 3.21, "theoretical_loss": 4.023882541840667, "tokens_seen": 389545984 }, { "epoch": 0.12, "learning_rate": 0.00044542627883650956, "loss": 3.2247, "theoretical_loss": 4.023806474062566, "tokens_seen": 389611520 }, { "epoch": 0.12, "learning_rate": 0.0004454162487462387, "loss": 2.8222, "theoretical_loss": 4.023730422660624, "tokens_seen": 389677056 }, { "epoch": 0.12, "learning_rate": 0.0004454062186559679, "loss": 3.1396, "theoretical_loss": 4.023654387628565, "tokens_seen": 389742592 }, { "epoch": 0.12, "learning_rate": 0.0004453961885656971, "loss": 3.0235, "theoretical_loss": 4.023578368960113, "tokens_seen": 389808128 }, { "epoch": 0.12, "learning_rate": 0.0004453861584754263, "loss": 2.8784, "theoretical_loss": 4.023502366648994, "tokens_seen": 389873664 }, { "epoch": 0.12, "objective/train/docs_used": 232298, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4993464946746826, "objective/train/theoretical_loss": 4.023426380688943, "objective/train/tokens_used": 410399200, "theoretical_loss": 4.023426380688943, "tokens_seen": 389939200 }, { "epoch": 0.12, "learning_rate": 0.00044537612838515547, "loss": 2.7087, "theoretical_loss": 4.023426380688943, "tokens_seen": 389939200 }, { "epoch": 0.12, "learning_rate": 0.00044536609829488465, "loss": 2.7692, "theoretical_loss": 4.023350411073693, "tokens_seen": 390004736 }, { "epoch": 0.12, "learning_rate": 0.00044535606820461383, "loss": 3.179, "theoretical_loss": 4.023274457796983, "tokens_seen": 390070272 }, { "epoch": 0.12, "learning_rate": 0.00044534603811434306, "loss": 3.0874, "theoretical_loss": 4.023198520852555, "tokens_seen": 390135808 }, { "epoch": 0.12, "learning_rate": 0.0004453360080240722, "loss": 2.9398, "theoretical_loss": 4.023122600234155, "tokens_seen": 390201344 }, { "epoch": 0.12, "learning_rate": 0.0004453259779338014, "loss": 2.7279, "theoretical_loss": 4.0230466959355295, "tokens_seen": 390266880 }, { "epoch": 0.12, "learning_rate": 0.00044531594784353055, "loss": 3.0187, "theoretical_loss": 4.022970807950434, "tokens_seen": 390332416 }, { "epoch": 0.12, "learning_rate": 0.0004453059177532598, "loss": 3.1262, "theoretical_loss": 4.022894936272621, "tokens_seen": 390397952 }, { "epoch": 0.12, "learning_rate": 0.00044529588766298897, "loss": 2.802, "theoretical_loss": 4.022819080895853, "tokens_seen": 390463488 }, { "epoch": 0.12, "learning_rate": 0.00044528585757271815, "loss": 3.0968, "theoretical_loss": 4.02274324181389, "tokens_seen": 390529024 }, { "epoch": 0.12, "learning_rate": 0.00044527582748244733, "loss": 2.9464, "theoretical_loss": 4.022667419020499, "tokens_seen": 390594560 }, { "epoch": 0.12, "learning_rate": 0.00044526579739217657, "loss": 2.8463, "theoretical_loss": 4.0225916125094505, "tokens_seen": 390660096 }, { "epoch": 0.12, "learning_rate": 0.0004452557673019057, "loss": 2.6302, "theoretical_loss": 4.022515822274515, "tokens_seen": 390725632 }, { "epoch": 0.12, "learning_rate": 0.00044524573721163493, "loss": 2.9971, "theoretical_loss": 4.0224400483094715, "tokens_seen": 390791168 }, { "epoch": 0.12, "learning_rate": 0.00044523570712136406, "loss": 2.9961, "theoretical_loss": 4.022364290608098, "tokens_seen": 390856704 }, { "epoch": 0.12, "learning_rate": 0.0004452256770310933, "loss": 3.0202, "theoretical_loss": 4.022288549164178, "tokens_seen": 390922240 }, { "epoch": 0.12, "learning_rate": 0.00044521564694082247, "loss": 2.9233, "theoretical_loss": 4.022212823971498, "tokens_seen": 390987776 }, { "epoch": 0.12, "learning_rate": 0.00044520561685055165, "loss": 3.165, "theoretical_loss": 4.022137115023848, "tokens_seen": 391053312 }, { "epoch": 0.12, "learning_rate": 0.00044519558676028083, "loss": 3.1339, "theoretical_loss": 4.022061422315023, "tokens_seen": 391118848 }, { "epoch": 0.12, "learning_rate": 0.00044518555667001, "loss": 2.9278, "theoretical_loss": 4.021985745838817, "tokens_seen": 391184384 }, { "epoch": 0.12, "learning_rate": 0.00044517552657973925, "loss": 2.7157, "theoretical_loss": 4.021910085589032, "tokens_seen": 391249920 }, { "epoch": 0.12, "learning_rate": 0.00044516549648946843, "loss": 2.8932, "theoretical_loss": 4.021834441559472, "tokens_seen": 391315456 }, { "epoch": 0.12, "learning_rate": 0.0004451554663991976, "loss": 2.9059, "theoretical_loss": 4.021758813743942, "tokens_seen": 391380992 }, { "epoch": 0.12, "learning_rate": 0.0004451454363089268, "loss": 2.9088, "theoretical_loss": 4.021683202136255, "tokens_seen": 391446528 }, { "epoch": 0.12, "learning_rate": 0.00044513540621865603, "loss": 2.995, "theoretical_loss": 4.021607606730223, "tokens_seen": 391512064 }, { "epoch": 0.12, "objective/train/docs_used": 233436, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.01359224319458, "objective/train/theoretical_loss": 4.0215320275196635, "objective/train/tokens_used": 412037600, "theoretical_loss": 4.0215320275196635, "tokens_seen": 391577600 }, { "epoch": 0.12, "learning_rate": 0.00044512537612838516, "loss": 2.9565, "theoretical_loss": 4.0215320275196635, "tokens_seen": 391577600 }, { "epoch": 0.12, "learning_rate": 0.0004451153460381144, "loss": 3.0462, "theoretical_loss": 4.0214564644983986, "tokens_seen": 391643136 }, { "epoch": 0.12, "learning_rate": 0.0004451053159478435, "loss": 2.6842, "theoretical_loss": 4.021380917660251, "tokens_seen": 391708672 }, { "epoch": 0.12, "learning_rate": 0.00044509528585757275, "loss": 3.0243, "theoretical_loss": 4.021305386999048, "tokens_seen": 391774208 }, { "epoch": 0.12, "learning_rate": 0.00044508525576730194, "loss": 2.9867, "theoretical_loss": 4.021229872508621, "tokens_seen": 391839744 }, { "epoch": 0.12, "learning_rate": 0.0004450752256770311, "loss": 2.6909, "theoretical_loss": 4.021154374182803, "tokens_seen": 391905280 }, { "epoch": 0.12, "learning_rate": 0.0004450651955867603, "loss": 2.7727, "theoretical_loss": 4.021078892015433, "tokens_seen": 391970816 }, { "epoch": 0.12, "learning_rate": 0.0004450551654964895, "loss": 3.0262, "theoretical_loss": 4.021003426000353, "tokens_seen": 392036352 }, { "epoch": 0.12, "learning_rate": 0.00044504513540621866, "loss": 3.0198, "theoretical_loss": 4.020927976131404, "tokens_seen": 392101888 }, { "epoch": 0.12, "learning_rate": 0.0004450351053159479, "loss": 3.085, "theoretical_loss": 4.0208525424024355, "tokens_seen": 392167424 }, { "epoch": 0.12, "learning_rate": 0.000445025075225677, "loss": 2.9415, "theoretical_loss": 4.020777124807299, "tokens_seen": 392232960 }, { "epoch": 0.12, "learning_rate": 0.00044501504513540626, "loss": 2.8681, "theoretical_loss": 4.020701723339848, "tokens_seen": 392298496 }, { "epoch": 0.12, "learning_rate": 0.0004450050150451354, "loss": 3.0426, "theoretical_loss": 4.020626337993941, "tokens_seen": 392364032 }, { "epoch": 0.12, "learning_rate": 0.0004449949849548646, "loss": 2.9429, "theoretical_loss": 4.020550968763439, "tokens_seen": 392429568 }, { "epoch": 0.12, "learning_rate": 0.0004449849548645938, "loss": 3.0646, "theoretical_loss": 4.020475615642207, "tokens_seen": 392495104 }, { "epoch": 0.12, "learning_rate": 0.000444974924774323, "loss": 2.7898, "theoretical_loss": 4.020400278624112, "tokens_seen": 392560640 }, { "epoch": 0.12, "learning_rate": 0.00044496489468405216, "loss": 2.9612, "theoretical_loss": 4.020324957703025, "tokens_seen": 392626176 }, { "epoch": 0.12, "learning_rate": 0.0004449548645937814, "loss": 2.9467, "theoretical_loss": 4.020249652872822, "tokens_seen": 392691712 }, { "epoch": 0.12, "learning_rate": 0.0004449448345035105, "loss": 2.6348, "theoretical_loss": 4.02017436412738, "tokens_seen": 392757248 }, { "epoch": 0.12, "learning_rate": 0.00044493480441323976, "loss": 3.0607, "theoretical_loss": 4.02009909146058, "tokens_seen": 392822784 }, { "epoch": 0.12, "learning_rate": 0.0004449247743229689, "loss": 2.826, "theoretical_loss": 4.020023834866308, "tokens_seen": 392888320 }, { "epoch": 0.12, "learning_rate": 0.0004449147442326981, "loss": 3.0724, "theoretical_loss": 4.01994859433845, "tokens_seen": 392953856 }, { "epoch": 0.12, "learning_rate": 0.0004449047141424273, "loss": 2.9021, "theoretical_loss": 4.019873369870899, "tokens_seen": 393019392 }, { "epoch": 0.12, "learning_rate": 0.0004448946840521565, "loss": 3.0023, "theoretical_loss": 4.019798161457548, "tokens_seen": 393084928 }, { "epoch": 0.12, "learning_rate": 0.00044488465396188567, "loss": 2.9141, "theoretical_loss": 4.019722969092298, "tokens_seen": 393150464 }, { "epoch": 0.12, "objective/train/docs_used": 234056, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1465675830841064, "objective/train/theoretical_loss": 4.019647792769048, "objective/train/tokens_used": 413676000, "theoretical_loss": 4.019647792769048, "tokens_seen": 393216000 }, { "epoch": 0.12, "learning_rate": 0.00044487462387161485, "loss": 3.207, "theoretical_loss": 4.019647792769048, "tokens_seen": 393216000 }, { "epoch": 0.12, "learning_rate": 0.00044486459378134403, "loss": 2.9752, "theoretical_loss": 4.019572632481702, "tokens_seen": 393281536 }, { "epoch": 0.12, "learning_rate": 0.00044485456369107326, "loss": 3.064, "theoretical_loss": 4.01949748822417, "tokens_seen": 393347072 }, { "epoch": 0.12, "learning_rate": 0.0004448445336008024, "loss": 2.7322, "theoretical_loss": 4.0194223599903625, "tokens_seen": 393412608 }, { "epoch": 0.12, "learning_rate": 0.0004448345035105316, "loss": 3.1126, "theoretical_loss": 4.019347247774195, "tokens_seen": 393478144 }, { "epoch": 0.12, "learning_rate": 0.00044482447342026075, "loss": 2.7526, "theoretical_loss": 4.0192721515695835, "tokens_seen": 393543680 }, { "epoch": 0.12, "learning_rate": 0.00044481444332999, "loss": 3.1153, "theoretical_loss": 4.019197071370451, "tokens_seen": 393609216 }, { "epoch": 0.12, "learning_rate": 0.00044480441323971917, "loss": 3.1239, "theoretical_loss": 4.019122007170722, "tokens_seen": 393674752 }, { "epoch": 0.12, "learning_rate": 0.00044479438314944835, "loss": 2.9529, "theoretical_loss": 4.019046958964324, "tokens_seen": 393740288 }, { "epoch": 0.12, "learning_rate": 0.00044478435305917753, "loss": 2.9906, "theoretical_loss": 4.018971926745189, "tokens_seen": 393805824 }, { "epoch": 0.12, "learning_rate": 0.00044477432296890677, "loss": 2.6571, "theoretical_loss": 4.01889691050725, "tokens_seen": 393871360 }, { "epoch": 0.12, "learning_rate": 0.0004447642928786359, "loss": 3.1611, "theoretical_loss": 4.018821910244446, "tokens_seen": 393936896 }, { "epoch": 0.12, "learning_rate": 0.00044475426278836513, "loss": 2.9492, "theoretical_loss": 4.018746925950719, "tokens_seen": 394002432 }, { "epoch": 0.12, "learning_rate": 0.00044474423269809426, "loss": 3.2417, "theoretical_loss": 4.018671957620012, "tokens_seen": 394067968 }, { "epoch": 0.12, "learning_rate": 0.0004447342026078235, "loss": 2.9681, "theoretical_loss": 4.018597005246273, "tokens_seen": 394133504 }, { "epoch": 0.12, "learning_rate": 0.00044472417251755267, "loss": 2.7605, "theoretical_loss": 4.018522068823454, "tokens_seen": 394199040 }, { "epoch": 0.12, "learning_rate": 0.00044471414242728185, "loss": 3.1142, "theoretical_loss": 4.01844714834551, "tokens_seen": 394264576 }, { "epoch": 0.12, "learning_rate": 0.00044470411233701103, "loss": 2.8985, "theoretical_loss": 4.018372243806397, "tokens_seen": 394330112 }, { "epoch": 0.12, "learning_rate": 0.0004446940822467402, "loss": 2.8504, "theoretical_loss": 4.018297355200076, "tokens_seen": 394395648 }, { "epoch": 0.12, "learning_rate": 0.0004446840521564694, "loss": 2.8258, "theoretical_loss": 4.018222482520512, "tokens_seen": 394461184 }, { "epoch": 0.12, "learning_rate": 0.00044467402206619863, "loss": 2.6849, "theoretical_loss": 4.018147625761673, "tokens_seen": 394526720 }, { "epoch": 0.12, "learning_rate": 0.00044466399197592776, "loss": 3.0831, "theoretical_loss": 4.018072784917529, "tokens_seen": 394592256 }, { "epoch": 0.12, "learning_rate": 0.000444653961885657, "loss": 3.0542, "theoretical_loss": 4.017997959982055, "tokens_seen": 394657792 }, { "epoch": 0.12, "learning_rate": 0.0004446439317953861, "loss": 2.948, "theoretical_loss": 4.017923150949228, "tokens_seen": 394723328 }, { "epoch": 0.12, "learning_rate": 0.00044463390170511536, "loss": 2.819, "theoretical_loss": 4.017848357813028, "tokens_seen": 394788864 }, { "epoch": 0.12, "objective/train/docs_used": 235431, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6532654762268066, "objective/train/theoretical_loss": 4.01777358056744, "objective/train/tokens_used": 415314400, "theoretical_loss": 4.01777358056744, "tokens_seen": 394854400 }, { "epoch": 0.12, "learning_rate": 0.00044462387161484454, "loss": 3.0346, "theoretical_loss": 4.01777358056744, "tokens_seen": 394854400 }, { "epoch": 0.12, "learning_rate": 0.0004446138415245737, "loss": 2.8124, "theoretical_loss": 4.017698819206451, "tokens_seen": 394919936 }, { "epoch": 0.12, "learning_rate": 0.0004446038114343029, "loss": 2.7088, "theoretical_loss": 4.017624073724052, "tokens_seen": 394985472 }, { "epoch": 0.12, "learning_rate": 0.00044459378134403214, "loss": 3.2092, "theoretical_loss": 4.017549344114236, "tokens_seen": 395051008 }, { "epoch": 0.12, "learning_rate": 0.00044458375125376126, "loss": 2.8705, "theoretical_loss": 4.017474630371, "tokens_seen": 395116544 }, { "epoch": 0.12, "learning_rate": 0.0004445737211634905, "loss": 3.0769, "theoretical_loss": 4.0173999324883445, "tokens_seen": 395182080 }, { "epoch": 0.12, "learning_rate": 0.0004445636910732196, "loss": 2.8085, "theoretical_loss": 4.0173252504602734, "tokens_seen": 395247616 }, { "epoch": 0.12, "learning_rate": 0.00044455366098294886, "loss": 2.7325, "theoretical_loss": 4.017250584280794, "tokens_seen": 395313152 }, { "epoch": 0.12, "learning_rate": 0.00044454363089267804, "loss": 2.9246, "theoretical_loss": 4.017175933943916, "tokens_seen": 395378688 }, { "epoch": 0.12, "learning_rate": 0.0004445336008024072, "loss": 2.6805, "theoretical_loss": 4.0171012994436515, "tokens_seen": 395444224 }, { "epoch": 0.12, "learning_rate": 0.0004445235707121364, "loss": 2.7515, "theoretical_loss": 4.017026680774019, "tokens_seen": 395509760 }, { "epoch": 0.12, "learning_rate": 0.0004445135406218656, "loss": 2.9577, "theoretical_loss": 4.016952077929037, "tokens_seen": 395575296 }, { "epoch": 0.12, "learning_rate": 0.00044450351053159477, "loss": 3.033, "theoretical_loss": 4.016877490902731, "tokens_seen": 395640832 }, { "epoch": 0.12, "learning_rate": 0.000444493480441324, "loss": 2.7227, "theoretical_loss": 4.016802919689124, "tokens_seen": 395706368 }, { "epoch": 0.12, "learning_rate": 0.00044448345035105313, "loss": 2.9002, "theoretical_loss": 4.0167283642822476, "tokens_seen": 395771904 }, { "epoch": 0.12, "learning_rate": 0.00044447342026078236, "loss": 3.0708, "theoretical_loss": 4.016653824676135, "tokens_seen": 395837440 }, { "epoch": 0.12, "learning_rate": 0.0004444633901705115, "loss": 2.9276, "theoretical_loss": 4.016579300864821, "tokens_seen": 395902976 }, { "epoch": 0.12, "learning_rate": 0.0004444533600802407, "loss": 2.8551, "theoretical_loss": 4.016504792842346, "tokens_seen": 395968512 }, { "epoch": 0.12, "learning_rate": 0.00044444332998996996, "loss": 2.9801, "theoretical_loss": 4.016430300602751, "tokens_seen": 396034048 }, { "epoch": 0.12, "learning_rate": 0.0004444332998996991, "loss": 2.949, "theoretical_loss": 4.016355824140085, "tokens_seen": 396099584 }, { "epoch": 0.12, "learning_rate": 0.0004444232698094283, "loss": 2.8753, "theoretical_loss": 4.016281363448394, "tokens_seen": 396165120 }, { "epoch": 0.12, "learning_rate": 0.0004444132397191575, "loss": 2.6945, "theoretical_loss": 4.016206918521731, "tokens_seen": 396230656 }, { "epoch": 0.12, "learning_rate": 0.0004444032096288867, "loss": 3.2419, "theoretical_loss": 4.016132489354153, "tokens_seen": 396296192 }, { "epoch": 0.12, "learning_rate": 0.00044439317953861587, "loss": 3.0525, "theoretical_loss": 4.016058075939716, "tokens_seen": 396361728 }, { "epoch": 0.12, "learning_rate": 0.00044438314944834505, "loss": 3.1025, "theoretical_loss": 4.015983678272484, "tokens_seen": 396427264 }, { "epoch": 0.12, "objective/train/docs_used": 235840, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4883968830108643, "objective/train/theoretical_loss": 4.015909296346521, "objective/train/tokens_used": 416952800, "theoretical_loss": 4.015909296346521, "tokens_seen": 396492800 }, { "epoch": 0.12, "learning_rate": 0.00044437311935807423, "loss": 2.8268, "theoretical_loss": 4.015909296346521, "tokens_seen": 396492800 }, { "epoch": 0.12, "learning_rate": 0.00044436308926780346, "loss": 2.8041, "theoretical_loss": 4.015834930155897, "tokens_seen": 396558336 }, { "epoch": 0.12, "learning_rate": 0.0004443530591775326, "loss": 2.8989, "theoretical_loss": 4.015760579694683, "tokens_seen": 396623872 }, { "epoch": 0.12, "learning_rate": 0.0004443430290872618, "loss": 3.0333, "theoretical_loss": 4.015686244956953, "tokens_seen": 396689408 }, { "epoch": 0.12, "learning_rate": 0.00044433299899699095, "loss": 2.93, "theoretical_loss": 4.015611925936785, "tokens_seen": 396754944 }, { "epoch": 0.12, "learning_rate": 0.0004443229689067202, "loss": 2.6344, "theoretical_loss": 4.01553762262826, "tokens_seen": 396820480 }, { "epoch": 0.12, "learning_rate": 0.00044431293881644937, "loss": 2.9469, "theoretical_loss": 4.015463335025465, "tokens_seen": 396886016 }, { "epoch": 0.12, "learning_rate": 0.00044430290872617855, "loss": 2.9232, "theoretical_loss": 4.015389063122484, "tokens_seen": 396951552 }, { "epoch": 0.12, "learning_rate": 0.00044429287863590773, "loss": 2.7601, "theoretical_loss": 4.015314806913411, "tokens_seen": 397017088 }, { "epoch": 0.12, "learning_rate": 0.00044428284854563697, "loss": 2.6964, "theoretical_loss": 4.015240566392338, "tokens_seen": 397082624 }, { "epoch": 0.12, "learning_rate": 0.0004442728184553661, "loss": 3.0921, "theoretical_loss": 4.015166341553362, "tokens_seen": 397148160 }, { "epoch": 0.12, "learning_rate": 0.00044426278836509533, "loss": 2.902, "theoretical_loss": 4.015092132390585, "tokens_seen": 397213696 }, { "epoch": 0.12, "learning_rate": 0.00044425275827482446, "loss": 2.7439, "theoretical_loss": 4.01501793889811, "tokens_seen": 397279232 }, { "epoch": 0.12, "learning_rate": 0.0004442427281845537, "loss": 2.9831, "theoretical_loss": 4.014943761070043, "tokens_seen": 397344768 }, { "epoch": 0.12, "learning_rate": 0.00044423269809428287, "loss": 2.841, "theoretical_loss": 4.014869598900496, "tokens_seen": 397410304 }, { "epoch": 0.12, "learning_rate": 0.00044422266800401205, "loss": 3.0263, "theoretical_loss": 4.01479545238358, "tokens_seen": 397475840 }, { "epoch": 0.12, "learning_rate": 0.00044421263791374123, "loss": 2.8788, "theoretical_loss": 4.014721321513412, "tokens_seen": 397541376 }, { "epoch": 0.12, "learning_rate": 0.0004442026078234704, "loss": 2.9554, "theoretical_loss": 4.0146472062841125, "tokens_seen": 397606912 }, { "epoch": 0.12, "learning_rate": 0.0004441925777331996, "loss": 2.8914, "theoretical_loss": 4.014573106689803, "tokens_seen": 397672448 }, { "epoch": 0.12, "learning_rate": 0.00044418254764292883, "loss": 2.9146, "theoretical_loss": 4.0144990227246105, "tokens_seen": 397737984 }, { "epoch": 0.12, "learning_rate": 0.00044417251755265796, "loss": 3.127, "theoretical_loss": 4.014424954382663, "tokens_seen": 397803520 }, { "epoch": 0.12, "learning_rate": 0.0004441624874623872, "loss": 3.1975, "theoretical_loss": 4.014350901658093, "tokens_seen": 397869056 }, { "epoch": 0.12, "learning_rate": 0.0004441524573721163, "loss": 2.8541, "theoretical_loss": 4.014276864545036, "tokens_seen": 397934592 }, { "epoch": 0.12, "learning_rate": 0.00044414242728184556, "loss": 3.0876, "theoretical_loss": 4.014202843037632, "tokens_seen": 398000128 }, { "epoch": 0.12, "learning_rate": 0.00044413239719157474, "loss": 2.8968, "theoretical_loss": 4.014128837130021, "tokens_seen": 398065664 }, { "epoch": 0.12, "objective/train/docs_used": 236468, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.918165445327759, "objective/train/theoretical_loss": 4.0140548468163475, "objective/train/tokens_used": 418591200, "theoretical_loss": 4.0140548468163475, "tokens_seen": 398131200 }, { "epoch": 0.12, "learning_rate": 0.0004441223671013039, "loss": 2.8995, "theoretical_loss": 4.0140548468163475, "tokens_seen": 398131200 }, { "epoch": 0.12, "learning_rate": 0.0004441123370110331, "loss": 3.0238, "theoretical_loss": 4.0139808720907615, "tokens_seen": 398196736 }, { "epoch": 0.12, "learning_rate": 0.00044410230692076234, "loss": 2.8314, "theoretical_loss": 4.013906912947412, "tokens_seen": 398262272 }, { "epoch": 0.12, "learning_rate": 0.00044409227683049146, "loss": 2.9981, "theoretical_loss": 4.013832969380456, "tokens_seen": 398327808 }, { "epoch": 0.12, "learning_rate": 0.0004440822467402207, "loss": 3.0402, "theoretical_loss": 4.013759041384049, "tokens_seen": 398393344 }, { "epoch": 0.12, "learning_rate": 0.0004440722166499498, "loss": 2.6356, "theoretical_loss": 4.013685128952353, "tokens_seen": 398458880 }, { "epoch": 0.12, "learning_rate": 0.00044406218655967906, "loss": 3.1591, "theoretical_loss": 4.013611232079531, "tokens_seen": 398524416 }, { "epoch": 0.12, "learning_rate": 0.00044405215646940824, "loss": 2.8642, "theoretical_loss": 4.01353735075975, "tokens_seen": 398589952 }, { "epoch": 0.12, "learning_rate": 0.0004440421263791374, "loss": 3.0194, "theoretical_loss": 4.01346348498718, "tokens_seen": 398655488 }, { "epoch": 0.12, "learning_rate": 0.0004440320962888666, "loss": 3.0007, "theoretical_loss": 4.013389634755995, "tokens_seen": 398721024 }, { "epoch": 0.12, "learning_rate": 0.0004440220661985958, "loss": 2.9131, "theoretical_loss": 4.013315800060371, "tokens_seen": 398786560 }, { "epoch": 0.12, "learning_rate": 0.00044401203610832497, "loss": 2.7356, "theoretical_loss": 4.013241980894488, "tokens_seen": 398852096 }, { "epoch": 0.12, "learning_rate": 0.0004440020060180542, "loss": 2.8804, "theoretical_loss": 4.013168177252528, "tokens_seen": 398917632 }, { "epoch": 0.12, "learning_rate": 0.00044399197592778333, "loss": 2.8778, "theoretical_loss": 4.013094389128677, "tokens_seen": 398983168 }, { "epoch": 0.12, "learning_rate": 0.00044398194583751256, "loss": 3.0424, "theoretical_loss": 4.0130206165171245, "tokens_seen": 399048704 }, { "epoch": 0.12, "learning_rate": 0.0004439719157472417, "loss": 3.0925, "theoretical_loss": 4.012946859412062, "tokens_seen": 399114240 }, { "epoch": 0.12, "learning_rate": 0.0004439618856569709, "loss": 2.771, "theoretical_loss": 4.012873117807685, "tokens_seen": 399179776 }, { "epoch": 0.12, "learning_rate": 0.0004439518555667001, "loss": 2.7898, "theoretical_loss": 4.012799391698191, "tokens_seen": 399245312 }, { "epoch": 0.12, "learning_rate": 0.0004439418254764293, "loss": 2.9576, "theoretical_loss": 4.012725681077783, "tokens_seen": 399310848 }, { "epoch": 0.12, "learning_rate": 0.00044393179538615847, "loss": 2.9726, "theoretical_loss": 4.012651985940666, "tokens_seen": 399376384 }, { "epoch": 0.12, "learning_rate": 0.0004439217652958877, "loss": 3.0252, "theoretical_loss": 4.012578306281045, "tokens_seen": 399441920 }, { "epoch": 0.12, "learning_rate": 0.00044391173520561683, "loss": 2.6768, "theoretical_loss": 4.012504642093133, "tokens_seen": 399507456 }, { "epoch": 0.12, "learning_rate": 0.00044390170511534607, "loss": 3.0128, "theoretical_loss": 4.012430993371144, "tokens_seen": 399572992 }, { "epoch": 0.12, "learning_rate": 0.0004438916750250752, "loss": 3.0036, "theoretical_loss": 4.012357360109293, "tokens_seen": 399638528 }, { "epoch": 0.12, "learning_rate": 0.00044388164493480443, "loss": 2.8824, "theoretical_loss": 4.012283742301802, "tokens_seen": 399704064 }, { "epoch": 0.12, "objective/train/docs_used": 237759, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.293104410171509, "objective/train/theoretical_loss": 4.012210139942894, "objective/train/tokens_used": 420229600, "theoretical_loss": 4.012210139942894, "tokens_seen": 399769600 }, { "epoch": 0.12, "learning_rate": 0.0004438716148445336, "loss": 3.0101, "theoretical_loss": 4.012210139942894, "tokens_seen": 399769600 }, { "epoch": 0.12, "learning_rate": 0.0004438615847542628, "loss": 2.7273, "theoretical_loss": 4.012136553026796, "tokens_seen": 399835136 }, { "epoch": 0.12, "learning_rate": 0.00044385155466399197, "loss": 2.9857, "theoretical_loss": 4.012062981547736, "tokens_seen": 399900672 }, { "epoch": 0.12, "learning_rate": 0.00044384152457372115, "loss": 2.9248, "theoretical_loss": 4.011989425499947, "tokens_seen": 399966208 }, { "epoch": 0.12, "learning_rate": 0.00044383149448345033, "loss": 2.9246, "theoretical_loss": 4.0119158848776655, "tokens_seen": 400031744 }, { "epoch": 0.12, "learning_rate": 0.00044382146439317957, "loss": 2.8758, "theoretical_loss": 4.011842359675129, "tokens_seen": 400097280 }, { "epoch": 0.12, "learning_rate": 0.0004438114343029087, "loss": 2.905, "theoretical_loss": 4.011768849886582, "tokens_seen": 400162816 }, { "epoch": 0.12, "learning_rate": 0.00044380140421263793, "loss": 2.9052, "theoretical_loss": 4.011695355506266, "tokens_seen": 400228352 }, { "epoch": 0.12, "learning_rate": 0.0004437913741223671, "loss": 3.0508, "theoretical_loss": 4.011621876528431, "tokens_seen": 400293888 }, { "epoch": 0.12, "learning_rate": 0.0004437813440320963, "loss": 2.9092, "theoretical_loss": 4.011548412947327, "tokens_seen": 400359424 }, { "epoch": 0.12, "learning_rate": 0.0004437713139418255, "loss": 2.6617, "theoretical_loss": 4.011474964757211, "tokens_seen": 400424960 }, { "epoch": 0.12, "learning_rate": 0.00044376128385155466, "loss": 2.9935, "theoretical_loss": 4.0114015319523375, "tokens_seen": 400490496 }, { "epoch": 0.12, "learning_rate": 0.00044375125376128384, "loss": 2.7608, "theoretical_loss": 4.011328114526968, "tokens_seen": 400556032 }, { "epoch": 0.12, "learning_rate": 0.00044374122367101307, "loss": 3.0189, "theoretical_loss": 4.011254712475365, "tokens_seen": 400621568 }, { "epoch": 0.12, "learning_rate": 0.0004437311935807422, "loss": 2.8798, "theoretical_loss": 4.011181325791797, "tokens_seen": 400687104 }, { "epoch": 0.12, "learning_rate": 0.00044372116349047143, "loss": 3.073, "theoretical_loss": 4.011107954470531, "tokens_seen": 400752640 }, { "epoch": 0.12, "learning_rate": 0.00044371113340020056, "loss": 3.0002, "theoretical_loss": 4.011034598505843, "tokens_seen": 400818176 }, { "epoch": 0.12, "learning_rate": 0.0004437011033099298, "loss": 2.6837, "theoretical_loss": 4.010961257892005, "tokens_seen": 400883712 }, { "epoch": 0.12, "learning_rate": 0.00044369107321965903, "loss": 3.0952, "theoretical_loss": 4.010887932623299, "tokens_seen": 400949248 }, { "epoch": 0.12, "learning_rate": 0.00044368104312938816, "loss": 3.1133, "theoretical_loss": 4.010814622694006, "tokens_seen": 401014784 }, { "epoch": 0.12, "learning_rate": 0.0004436710130391174, "loss": 2.6992, "theoretical_loss": 4.01074132809841, "tokens_seen": 401080320 }, { "epoch": 0.12, "learning_rate": 0.0004436609829488465, "loss": 3.0919, "theoretical_loss": 4.0106680488308, "tokens_seen": 401145856 }, { "epoch": 0.12, "learning_rate": 0.00044365095285857576, "loss": 2.9937, "theoretical_loss": 4.010594784885466, "tokens_seen": 401211392 }, { "epoch": 0.12, "learning_rate": 0.00044364092276830494, "loss": 2.6733, "theoretical_loss": 4.010521536256705, "tokens_seen": 401276928 }, { "epoch": 0.12, "learning_rate": 0.0004436308926780341, "loss": 2.933, "theoretical_loss": 4.01044830293881, "tokens_seen": 401342464 }, { "epoch": 0.12, "objective/train/docs_used": 238664, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.03971004486084, "objective/train/theoretical_loss": 4.0103750849260855, "objective/train/tokens_used": 421868000, "theoretical_loss": 4.0103750849260855, "tokens_seen": 401408000 }, { "epoch": 0.12, "learning_rate": 0.0004436208625877633, "loss": 2.7744, "theoretical_loss": 4.0103750849260855, "tokens_seen": 401408000 }, { "epoch": 0.12, "learning_rate": 0.00044361083249749254, "loss": 2.9641, "theoretical_loss": 4.010301882212831, "tokens_seen": 401473536 }, { "epoch": 0.12, "learning_rate": 0.00044360080240722166, "loss": 2.9476, "theoretical_loss": 4.010228694793356, "tokens_seen": 401539072 }, { "epoch": 0.12, "learning_rate": 0.0004435907723169509, "loss": 3.1747, "theoretical_loss": 4.010155522661968, "tokens_seen": 401604608 }, { "epoch": 0.12, "learning_rate": 0.00044358074222668, "loss": 2.8852, "theoretical_loss": 4.0100823658129805, "tokens_seen": 401670144 }, { "epoch": 0.12, "learning_rate": 0.00044357071213640926, "loss": 3.0247, "theoretical_loss": 4.010009224240708, "tokens_seen": 401735680 }, { "epoch": 0.12, "learning_rate": 0.00044356068204613844, "loss": 2.9916, "theoretical_loss": 4.0099360979394705, "tokens_seen": 401801216 }, { "epoch": 0.12, "learning_rate": 0.0004435506519558676, "loss": 2.9303, "theoretical_loss": 4.009862986903588, "tokens_seen": 401866752 }, { "epoch": 0.12, "learning_rate": 0.0004435406218655968, "loss": 2.8707, "theoretical_loss": 4.009789891127387, "tokens_seen": 401932288 }, { "epoch": 0.12, "learning_rate": 0.000443530591775326, "loss": 2.9204, "theoretical_loss": 4.009716810605193, "tokens_seen": 401997824 }, { "epoch": 0.12, "learning_rate": 0.00044352056168505517, "loss": 2.7006, "theoretical_loss": 4.009643745331338, "tokens_seen": 402063360 }, { "epoch": 0.12, "learning_rate": 0.0004435105315947844, "loss": 2.9334, "theoretical_loss": 4.009570695300155, "tokens_seen": 402128896 }, { "epoch": 0.12, "learning_rate": 0.00044350050150451353, "loss": 2.986, "theoretical_loss": 4.009497660505982, "tokens_seen": 402194432 }, { "epoch": 0.12, "learning_rate": 0.00044349047141424276, "loss": 2.9527, "theoretical_loss": 4.009424640943157, "tokens_seen": 402259968 }, { "epoch": 0.12, "learning_rate": 0.0004434804413239719, "loss": 3.0015, "theoretical_loss": 4.009351636606025, "tokens_seen": 402325504 }, { "epoch": 0.12, "learning_rate": 0.0004434704112337011, "loss": 2.7869, "theoretical_loss": 4.00927864748893, "tokens_seen": 402391040 }, { "epoch": 0.12, "learning_rate": 0.0004434603811434303, "loss": 2.8129, "theoretical_loss": 4.009205673586222, "tokens_seen": 402456576 }, { "epoch": 0.12, "learning_rate": 0.0004434503510531595, "loss": 2.8991, "theoretical_loss": 4.009132714892251, "tokens_seen": 402522112 }, { "epoch": 0.12, "learning_rate": 0.00044344032096288867, "loss": 2.8425, "theoretical_loss": 4.009059771401375, "tokens_seen": 402587648 }, { "epoch": 0.12, "learning_rate": 0.0004434302908726179, "loss": 2.8532, "theoretical_loss": 4.008986843107949, "tokens_seen": 402653184 }, { "epoch": 0.12, "learning_rate": 0.00044342026078234703, "loss": 2.798, "theoretical_loss": 4.008913930006336, "tokens_seen": 402718720 }, { "epoch": 0.12, "learning_rate": 0.00044341023069207627, "loss": 2.9858, "theoretical_loss": 4.008841032090898, "tokens_seen": 402784256 }, { "epoch": 0.12, "learning_rate": 0.0004434002006018054, "loss": 3.017, "theoretical_loss": 4.008768149356003, "tokens_seen": 402849792 }, { "epoch": 0.12, "learning_rate": 0.00044339017051153463, "loss": 2.7734, "theoretical_loss": 4.0086952817960215, "tokens_seen": 402915328 }, { "epoch": 0.12, "learning_rate": 0.0004433801404212638, "loss": 3.0587, "theoretical_loss": 4.0086224294053245, "tokens_seen": 402980864 }, { "epoch": 0.12, "objective/train/docs_used": 240054, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.683651924133301, "objective/train/theoretical_loss": 4.008549592178291, "objective/train/tokens_used": 423506400, "theoretical_loss": 4.008549592178291, "tokens_seen": 403046400 }, { "epoch": 0.12, "learning_rate": 0.000443370110330993, "loss": 2.7831, "theoretical_loss": 4.008549592178291, "tokens_seen": 403046400 }, { "epoch": 0.12, "learning_rate": 0.00044336008024072217, "loss": 3.0083, "theoretical_loss": 4.008476770109296, "tokens_seen": 403111936 }, { "epoch": 0.12, "learning_rate": 0.00044335005015045135, "loss": 2.8457, "theoretical_loss": 4.008403963192725, "tokens_seen": 403177472 }, { "epoch": 0.12, "learning_rate": 0.00044334002006018053, "loss": 3.0476, "theoretical_loss": 4.008331171422961, "tokens_seen": 403243008 }, { "epoch": 0.12, "learning_rate": 0.00044332998996990977, "loss": 2.9666, "theoretical_loss": 4.008258394794392, "tokens_seen": 403308544 }, { "epoch": 0.12, "learning_rate": 0.0004433199598796389, "loss": 2.9, "theoretical_loss": 4.0081856333014105, "tokens_seen": 403374080 }, { "epoch": 0.12, "learning_rate": 0.00044330992978936813, "loss": 2.8943, "theoretical_loss": 4.008112886938409, "tokens_seen": 403439616 }, { "epoch": 0.12, "learning_rate": 0.0004432998996990973, "loss": 2.8262, "theoretical_loss": 4.008040155699783, "tokens_seen": 403505152 }, { "epoch": 0.12, "learning_rate": 0.0004432898696088265, "loss": 2.9041, "theoretical_loss": 4.007967439579936, "tokens_seen": 403570688 }, { "epoch": 0.12, "learning_rate": 0.0004432798395185557, "loss": 3.088, "theoretical_loss": 4.007894738573267, "tokens_seen": 403636224 }, { "epoch": 0.12, "learning_rate": 0.00044326980942828486, "loss": 2.7578, "theoretical_loss": 4.007822052674185, "tokens_seen": 403701760 }, { "epoch": 0.12, "learning_rate": 0.00044325977933801404, "loss": 2.875, "theoretical_loss": 4.007749381877097, "tokens_seen": 403767296 }, { "epoch": 0.12, "learning_rate": 0.00044324974924774327, "loss": 3.1834, "theoretical_loss": 4.007676726176415, "tokens_seen": 403832832 }, { "epoch": 0.12, "learning_rate": 0.0004432397191574724, "loss": 2.5301, "theoretical_loss": 4.0076040855665545, "tokens_seen": 403898368 }, { "epoch": 0.12, "learning_rate": 0.00044322968906720163, "loss": 2.9509, "theoretical_loss": 4.007531460041933, "tokens_seen": 403963904 }, { "epoch": 0.12, "learning_rate": 0.00044321965897693076, "loss": 2.6442, "theoretical_loss": 4.00745884959697, "tokens_seen": 404029440 }, { "epoch": 0.12, "learning_rate": 0.00044320962888666, "loss": 3.2275, "theoretical_loss": 4.00738625422609, "tokens_seen": 404094976 }, { "epoch": 0.12, "learning_rate": 0.0004431995987963892, "loss": 2.8503, "theoretical_loss": 4.00731367392372, "tokens_seen": 404160512 }, { "epoch": 0.12, "learning_rate": 0.00044318956870611836, "loss": 3.0621, "theoretical_loss": 4.0072411086842905, "tokens_seen": 404226048 }, { "epoch": 0.12, "learning_rate": 0.00044317953861584754, "loss": 3.1279, "theoretical_loss": 4.007168558502232, "tokens_seen": 404291584 }, { "epoch": 0.12, "learning_rate": 0.0004431695085255767, "loss": 2.9978, "theoretical_loss": 4.007096023371981, "tokens_seen": 404357120 }, { "epoch": 0.12, "learning_rate": 0.0004431594784353059, "loss": 2.9717, "theoretical_loss": 4.007023503287976, "tokens_seen": 404422656 }, { "epoch": 0.12, "learning_rate": 0.00044314944834503514, "loss": 3.0025, "theoretical_loss": 4.006950998244658, "tokens_seen": 404488192 }, { "epoch": 0.12, "learning_rate": 0.00044313941825476427, "loss": 2.7928, "theoretical_loss": 4.006878508236473, "tokens_seen": 404553728 }, { "epoch": 0.12, "learning_rate": 0.0004431293881644935, "loss": 3.1174, "theoretical_loss": 4.006806033257866, "tokens_seen": 404619264 }, { "epoch": 0.12, "objective/train/docs_used": 240788, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.887927770614624, "objective/train/theoretical_loss": 4.006733573303289, "objective/train/tokens_used": 425144800, "theoretical_loss": 4.006733573303289, "tokens_seen": 404684800 }, { "epoch": 0.12, "learning_rate": 0.0004431193580742227, "loss": 2.8302, "theoretical_loss": 4.006733573303289, "tokens_seen": 404684800 }, { "epoch": 0.12, "learning_rate": 0.00044310932798395186, "loss": 3.0272, "theoretical_loss": 4.0066611283671945, "tokens_seen": 404750336 }, { "epoch": 0.12, "learning_rate": 0.00044309929789368104, "loss": 2.8301, "theoretical_loss": 4.0065886984440375, "tokens_seen": 404815872 }, { "epoch": 0.12, "learning_rate": 0.0004430892678034102, "loss": 3.0603, "theoretical_loss": 4.00651628352828, "tokens_seen": 404881408 }, { "epoch": 0.12, "learning_rate": 0.0004430792377131394, "loss": 2.8986, "theoretical_loss": 4.0064438836143825, "tokens_seen": 404946944 }, { "epoch": 0.12, "learning_rate": 0.00044306920762286864, "loss": 3.0905, "theoretical_loss": 4.00637149869681, "tokens_seen": 405012480 }, { "epoch": 0.12, "learning_rate": 0.00044305917753259777, "loss": 3.0358, "theoretical_loss": 4.0062991287700305, "tokens_seen": 405078016 }, { "epoch": 0.12, "learning_rate": 0.000443049147442327, "loss": 2.8933, "theoretical_loss": 4.006226773828514, "tokens_seen": 405143552 }, { "epoch": 0.12, "learning_rate": 0.00044303911735205613, "loss": 2.9264, "theoretical_loss": 4.006154433866736, "tokens_seen": 405209088 }, { "epoch": 0.12, "learning_rate": 0.00044302908726178537, "loss": 3.1129, "theoretical_loss": 4.0060821088791725, "tokens_seen": 405274624 }, { "epoch": 0.12, "learning_rate": 0.00044301905717151455, "loss": 2.9723, "theoretical_loss": 4.006009798860303, "tokens_seen": 405340160 }, { "epoch": 0.12, "learning_rate": 0.00044300902708124373, "loss": 2.9772, "theoretical_loss": 4.00593750380461, "tokens_seen": 405405696 }, { "epoch": 0.12, "learning_rate": 0.0004429989969909729, "loss": 3.0805, "theoretical_loss": 4.005865223706579, "tokens_seen": 405471232 }, { "epoch": 0.12, "learning_rate": 0.0004429889669007021, "loss": 3.0728, "theoretical_loss": 4.005792958560699, "tokens_seen": 405536768 }, { "epoch": 0.12, "learning_rate": 0.00044297893681043127, "loss": 2.835, "theoretical_loss": 4.005720708361461, "tokens_seen": 405602304 }, { "epoch": 0.12, "learning_rate": 0.0004429689067201605, "loss": 3.1059, "theoretical_loss": 4.00564847310336, "tokens_seen": 405667840 }, { "epoch": 0.12, "learning_rate": 0.00044295887662988963, "loss": 2.8118, "theoretical_loss": 4.005576252780892, "tokens_seen": 405733376 }, { "epoch": 0.12, "learning_rate": 0.00044294884653961887, "loss": 2.815, "theoretical_loss": 4.005504047388558, "tokens_seen": 405798912 }, { "epoch": 0.12, "learning_rate": 0.0004429388164493481, "loss": 2.8697, "theoretical_loss": 4.00543185692086, "tokens_seen": 405864448 }, { "epoch": 0.12, "learning_rate": 0.00044292878635907723, "loss": 2.9243, "theoretical_loss": 4.005359681372306, "tokens_seen": 405929984 }, { "epoch": 0.12, "learning_rate": 0.00044291875626880647, "loss": 2.9472, "theoretical_loss": 4.005287520737403, "tokens_seen": 405995520 }, { "epoch": 0.12, "learning_rate": 0.0004429087261785356, "loss": 2.8168, "theoretical_loss": 4.005215375010662, "tokens_seen": 406061056 }, { "epoch": 0.12, "learning_rate": 0.00044289869608826483, "loss": 3.1313, "theoretical_loss": 4.0051432441866, "tokens_seen": 406126592 }, { "epoch": 0.12, "learning_rate": 0.000442888665997994, "loss": 2.6992, "theoretical_loss": 4.005071128259734, "tokens_seen": 406192128 }, { "epoch": 0.12, "learning_rate": 0.0004428786359077232, "loss": 2.9379, "theoretical_loss": 4.004999027224584, "tokens_seen": 406257664 }, { "epoch": 0.12, "objective/train/docs_used": 241796, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5601613521575928, "objective/train/theoretical_loss": 4.004926941075674, "objective/train/tokens_used": 426783200, "theoretical_loss": 4.004926941075674, "tokens_seen": 406323200 }, { "epoch": 0.12, "learning_rate": 0.00044286860581745237, "loss": 2.7494, "theoretical_loss": 4.004926941075674, "tokens_seen": 406323200 }, { "epoch": 0.12, "learning_rate": 0.00044285857572718155, "loss": 2.8608, "theoretical_loss": 4.004854869807528, "tokens_seen": 406388736 }, { "epoch": 0.12, "learning_rate": 0.00044284854563691073, "loss": 3.0428, "theoretical_loss": 4.004782813414678, "tokens_seen": 406454272 }, { "epoch": 0.12, "learning_rate": 0.00044283851554663997, "loss": 2.8653, "theoretical_loss": 4.004710771891654, "tokens_seen": 406519808 }, { "epoch": 0.12, "learning_rate": 0.0004428284854563691, "loss": 3.0127, "theoretical_loss": 4.0046387452329935, "tokens_seen": 406585344 }, { "epoch": 0.12, "learning_rate": 0.00044281845536609833, "loss": 2.8257, "theoretical_loss": 4.004566733433231, "tokens_seen": 406650880 }, { "epoch": 0.12, "learning_rate": 0.0004428084252758275, "loss": 2.8753, "theoretical_loss": 4.00449473648691, "tokens_seen": 406716416 }, { "epoch": 0.12, "learning_rate": 0.0004427983951855567, "loss": 2.8526, "theoretical_loss": 4.004422754388573, "tokens_seen": 406781952 }, { "epoch": 0.12, "learning_rate": 0.0004427883650952859, "loss": 2.9927, "theoretical_loss": 4.004350787132766, "tokens_seen": 406847488 }, { "epoch": 0.12, "learning_rate": 0.00044277833500501506, "loss": 2.7786, "theoretical_loss": 4.004278834714039, "tokens_seen": 406913024 }, { "epoch": 0.12, "learning_rate": 0.00044276830491474424, "loss": 2.8439, "theoretical_loss": 4.004206897126943, "tokens_seen": 406978560 }, { "epoch": 0.12, "learning_rate": 0.00044275827482447347, "loss": 2.8901, "theoretical_loss": 4.004134974366036, "tokens_seen": 407044096 }, { "epoch": 0.12, "learning_rate": 0.0004427482447342026, "loss": 3.004, "theoretical_loss": 4.004063066425874, "tokens_seen": 407109632 }, { "epoch": 0.12, "learning_rate": 0.00044273821464393184, "loss": 2.8771, "theoretical_loss": 4.003991173301018, "tokens_seen": 407175168 }, { "epoch": 0.12, "learning_rate": 0.00044272818455366096, "loss": 2.9344, "theoretical_loss": 4.003919294986032, "tokens_seen": 407240704 }, { "epoch": 0.12, "learning_rate": 0.0004427181544633902, "loss": 2.919, "theoretical_loss": 4.003847431475482, "tokens_seen": 407306240 }, { "epoch": 0.12, "learning_rate": 0.0004427081243731194, "loss": 2.9829, "theoretical_loss": 4.003775582763939, "tokens_seen": 407371776 }, { "epoch": 0.12, "learning_rate": 0.00044269809428284856, "loss": 2.9322, "theoretical_loss": 4.0037037488459735, "tokens_seen": 407437312 }, { "epoch": 0.12, "learning_rate": 0.00044268806419257774, "loss": 2.9934, "theoretical_loss": 4.003631929716162, "tokens_seen": 407502848 }, { "epoch": 0.12, "learning_rate": 0.0004426780341023069, "loss": 3.0835, "theoretical_loss": 4.0035601253690825, "tokens_seen": 407568384 }, { "epoch": 0.12, "learning_rate": 0.0004426680040120361, "loss": 3.2281, "theoretical_loss": 4.0034883357993145, "tokens_seen": 407633920 }, { "epoch": 0.12, "learning_rate": 0.00044265797392176534, "loss": 3.0159, "theoretical_loss": 4.003416561001444, "tokens_seen": 407699456 }, { "epoch": 0.12, "learning_rate": 0.00044264794383149447, "loss": 3.0396, "theoretical_loss": 4.003344800970056, "tokens_seen": 407764992 }, { "epoch": 0.12, "learning_rate": 0.0004426379137412237, "loss": 3.1254, "theoretical_loss": 4.003273055699741, "tokens_seen": 407830528 }, { "epoch": 0.12, "learning_rate": 0.0004426278836509529, "loss": 2.7877, "theoretical_loss": 4.00320132518509, "tokens_seen": 407896064 }, { "epoch": 0.12, "objective/train/docs_used": 242638, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4354865550994873, "objective/train/theoretical_loss": 4.0031296094207, "objective/train/tokens_used": 428421600, "theoretical_loss": 4.0031296094207, "tokens_seen": 407961600 }, { "epoch": 0.12, "learning_rate": 0.00044261785356068206, "loss": 2.9397, "theoretical_loss": 4.0031296094207, "tokens_seen": 407961600 }, { "epoch": 0.12, "learning_rate": 0.00044260782347041124, "loss": 2.9511, "theoretical_loss": 4.003057908401168, "tokens_seen": 408027136 }, { "epoch": 0.12, "learning_rate": 0.0004425977933801404, "loss": 2.7822, "theoretical_loss": 4.002986222121095, "tokens_seen": 408092672 }, { "epoch": 0.12, "learning_rate": 0.0004425877632898696, "loss": 3.104, "theoretical_loss": 4.002914550575085, "tokens_seen": 408158208 }, { "epoch": 0.12, "learning_rate": 0.00044257773319959884, "loss": 3.1355, "theoretical_loss": 4.002842893757744, "tokens_seen": 408223744 }, { "epoch": 0.12, "learning_rate": 0.00044256770310932797, "loss": 2.9234, "theoretical_loss": 4.002771251663681, "tokens_seen": 408289280 }, { "epoch": 0.12, "learning_rate": 0.0004425576730190572, "loss": 2.8301, "theoretical_loss": 4.00269962428751, "tokens_seen": 408354816 }, { "epoch": 0.12, "learning_rate": 0.00044254764292878633, "loss": 3.0186, "theoretical_loss": 4.0026280116238455, "tokens_seen": 408420352 }, { "epoch": 0.12, "learning_rate": 0.00044253761283851557, "loss": 3.0234, "theoretical_loss": 4.002556413667305, "tokens_seen": 408485888 }, { "epoch": 0.12, "learning_rate": 0.00044252758274824475, "loss": 2.8821, "theoretical_loss": 4.0024848304125085, "tokens_seen": 408551424 }, { "epoch": 0.12, "learning_rate": 0.00044251755265797393, "loss": 2.8889, "theoretical_loss": 4.002413261854081, "tokens_seen": 408616960 }, { "epoch": 0.12, "learning_rate": 0.0004425075225677031, "loss": 2.8424, "theoretical_loss": 4.002341707986648, "tokens_seen": 408682496 }, { "epoch": 0.12, "learning_rate": 0.0004424974924774323, "loss": 3.0701, "theoretical_loss": 4.00227016880484, "tokens_seen": 408748032 }, { "epoch": 0.12, "learning_rate": 0.00044248746238716147, "loss": 2.7459, "theoretical_loss": 4.002198644303288, "tokens_seen": 408813568 }, { "epoch": 0.12, "learning_rate": 0.0004424774322968907, "loss": 2.9571, "theoretical_loss": 4.002127134476626, "tokens_seen": 408879104 }, { "epoch": 0.12, "learning_rate": 0.00044246740220661983, "loss": 2.8098, "theoretical_loss": 4.002055639319494, "tokens_seen": 408944640 }, { "epoch": 0.12, "learning_rate": 0.00044245737211634907, "loss": 2.9085, "theoretical_loss": 4.00198415882653, "tokens_seen": 409010176 }, { "epoch": 0.12, "learning_rate": 0.00044244734202607825, "loss": 3.1644, "theoretical_loss": 4.001912692992379, "tokens_seen": 409075712 }, { "epoch": 0.12, "learning_rate": 0.00044243731193580743, "loss": 3.0164, "theoretical_loss": 4.001841241811688, "tokens_seen": 409141248 }, { "epoch": 0.12, "learning_rate": 0.0004424272818455366, "loss": 2.7907, "theoretical_loss": 4.001769805279103, "tokens_seen": 409206784 }, { "epoch": 0.12, "learning_rate": 0.0004424172517552658, "loss": 2.9686, "theoretical_loss": 4.001698383389279, "tokens_seen": 409272320 }, { "epoch": 0.12, "learning_rate": 0.000442407221664995, "loss": 2.8624, "theoretical_loss": 4.001626976136868, "tokens_seen": 409337856 }, { "epoch": 0.12, "learning_rate": 0.0004423971915747242, "loss": 3.0741, "theoretical_loss": 4.001555583516529, "tokens_seen": 409403392 }, { "epoch": 0.12, "learning_rate": 0.00044238716148445334, "loss": 3.0561, "theoretical_loss": 4.001484205522922, "tokens_seen": 409468928 }, { "epoch": 0.12, "learning_rate": 0.00044237713139418257, "loss": 2.9995, "theoretical_loss": 4.00141284215071, "tokens_seen": 409534464 }, { "epoch": 0.12, "objective/train/docs_used": 244086, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1447839736938477, "objective/train/theoretical_loss": 4.001341493394558, "objective/train/tokens_used": 430060000, "theoretical_loss": 4.001341493394558, "tokens_seen": 409600000 }, { "epoch": 0.12, "learning_rate": 0.0004423671013039117, "loss": 2.9803, "theoretical_loss": 4.001341493394558, "tokens_seen": 409600000 }, { "epoch": 0.12, "learning_rate": 0.00044235707121364093, "loss": 2.9221, "theoretical_loss": 4.0012701592491355, "tokens_seen": 409665536 }, { "epoch": 0.12, "learning_rate": 0.0004423470411233701, "loss": 3.0829, "theoretical_loss": 4.001198839709113, "tokens_seen": 409731072 }, { "epoch": 0.12, "learning_rate": 0.0004423370110330993, "loss": 2.9286, "theoretical_loss": 4.001127534769166, "tokens_seen": 409796608 }, { "epoch": 0.12, "learning_rate": 0.0004423269809428285, "loss": 3.0952, "theoretical_loss": 4.001056244423971, "tokens_seen": 409862144 }, { "epoch": 0.12, "learning_rate": 0.0004423169508525577, "loss": 2.9253, "theoretical_loss": 4.000984968668206, "tokens_seen": 409927680 }, { "epoch": 0.12, "learning_rate": 0.00044230692076228684, "loss": 2.7064, "theoretical_loss": 4.000913707496556, "tokens_seen": 409993216 }, { "epoch": 0.12, "learning_rate": 0.0004422968906720161, "loss": 2.9732, "theoretical_loss": 4.000842460903706, "tokens_seen": 410058752 }, { "epoch": 0.12, "learning_rate": 0.0004422868605817452, "loss": 3.0309, "theoretical_loss": 4.000771228884344, "tokens_seen": 410124288 }, { "epoch": 0.12, "learning_rate": 0.00044227683049147444, "loss": 2.8903, "theoretical_loss": 4.0007000114331595, "tokens_seen": 410189824 }, { "epoch": 0.12, "learning_rate": 0.0004422668004012036, "loss": 2.9111, "theoretical_loss": 4.000628808544848, "tokens_seen": 410255360 }, { "epoch": 0.12, "learning_rate": 0.0004422567703109328, "loss": 3.1261, "theoretical_loss": 4.000557620214105, "tokens_seen": 410320896 }, { "epoch": 0.12, "learning_rate": 0.000442246740220662, "loss": 3.0598, "theoretical_loss": 4.000486446435629, "tokens_seen": 410386432 }, { "epoch": 0.12, "learning_rate": 0.00044223671013039116, "loss": 2.9045, "theoretical_loss": 4.000415287204124, "tokens_seen": 410451968 }, { "epoch": 0.12, "learning_rate": 0.00044222668004012034, "loss": 2.5777, "theoretical_loss": 4.000344142514294, "tokens_seen": 410517504 }, { "epoch": 0.12, "learning_rate": 0.0004422166499498496, "loss": 2.847, "theoretical_loss": 4.0002730123608465, "tokens_seen": 410583040 }, { "epoch": 0.12, "learning_rate": 0.0004422066198595787, "loss": 2.9707, "theoretical_loss": 4.000201896738492, "tokens_seen": 410648576 }, { "epoch": 0.12, "learning_rate": 0.00044219658976930794, "loss": 2.8757, "theoretical_loss": 4.000130795641944, "tokens_seen": 410714112 }, { "epoch": 0.12, "learning_rate": 0.0004421865596790371, "loss": 2.6645, "theoretical_loss": 4.0000597090659165, "tokens_seen": 410779648 }, { "epoch": 0.12, "learning_rate": 0.0004421765295887663, "loss": 2.6402, "theoretical_loss": 3.9999886370051305, "tokens_seen": 410845184 }, { "epoch": 0.12, "learning_rate": 0.00044216649949849554, "loss": 2.8724, "theoretical_loss": 3.9999175794543063, "tokens_seen": 410910720 }, { "epoch": 0.12, "learning_rate": 0.00044215646940822467, "loss": 2.8293, "theoretical_loss": 3.9998465364081683, "tokens_seen": 410976256 }, { "epoch": 0.12, "learning_rate": 0.0004421464393179539, "loss": 2.8997, "theoretical_loss": 3.999775507861443, "tokens_seen": 411041792 }, { "epoch": 0.12, "learning_rate": 0.0004421364092276831, "loss": 2.8315, "theoretical_loss": 3.999704493808861, "tokens_seen": 411107328 }, { "epoch": 0.12, "learning_rate": 0.00044212637913741226, "loss": 3.1415, "theoretical_loss": 3.9996334942451544, "tokens_seen": 411172864 }, { "epoch": 0.12, "objective/train/docs_used": 244630, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0692129135131836, "objective/train/theoretical_loss": 3.999562509165058, "objective/train/tokens_used": 431698400, "theoretical_loss": 3.999562509165058, "tokens_seen": 411238400 }, { "epoch": 0.12, "learning_rate": 0.00044211634904714144, "loss": 3.1408, "theoretical_loss": 3.999562509165058, "tokens_seen": 411238400 }, { "epoch": 0.12, "learning_rate": 0.0004421063189568706, "loss": 2.9589, "theoretical_loss": 3.9994915385633103, "tokens_seen": 411303936 }, { "epoch": 0.12, "learning_rate": 0.0004420962888665998, "loss": 3.0151, "theoretical_loss": 3.9994205824346514, "tokens_seen": 411369472 }, { "epoch": 0.12, "learning_rate": 0.00044208625877632904, "loss": 3.0807, "theoretical_loss": 3.9993496407738256, "tokens_seen": 411435008 }, { "epoch": 0.12, "learning_rate": 0.00044207622868605817, "loss": 2.8219, "theoretical_loss": 3.9992787135755785, "tokens_seen": 411500544 }, { "epoch": 0.12, "learning_rate": 0.0004420661985957874, "loss": 2.8975, "theoretical_loss": 3.999207800834659, "tokens_seen": 411566080 }, { "epoch": 0.12, "learning_rate": 0.00044205616850551653, "loss": 2.7822, "theoretical_loss": 3.9991369025458194, "tokens_seen": 411631616 }, { "epoch": 0.12, "learning_rate": 0.00044204613841524577, "loss": 2.8842, "theoretical_loss": 3.999066018703814, "tokens_seen": 411697152 }, { "epoch": 0.12, "learning_rate": 0.00044203610832497495, "loss": 2.8709, "theoretical_loss": 3.9989951493033997, "tokens_seen": 411762688 }, { "epoch": 0.12, "learning_rate": 0.00044202607823470413, "loss": 3.0365, "theoretical_loss": 3.9989242943393357, "tokens_seen": 411828224 }, { "epoch": 0.12, "learning_rate": 0.0004420160481444333, "loss": 2.9622, "theoretical_loss": 3.998853453806386, "tokens_seen": 411893760 }, { "epoch": 0.12, "learning_rate": 0.0004420060180541625, "loss": 2.8304, "theoretical_loss": 3.998782627699315, "tokens_seen": 411959296 }, { "epoch": 0.12, "learning_rate": 0.00044199598796389167, "loss": 3.0473, "theoretical_loss": 3.998711816012891, "tokens_seen": 412024832 }, { "epoch": 0.12, "learning_rate": 0.0004419859578736209, "loss": 2.8868, "theoretical_loss": 3.9986410187418855, "tokens_seen": 412090368 }, { "epoch": 0.12, "learning_rate": 0.00044197592778335003, "loss": 2.9131, "theoretical_loss": 3.9985702358810715, "tokens_seen": 412155904 }, { "epoch": 0.12, "learning_rate": 0.00044196589769307927, "loss": 2.8187, "theoretical_loss": 3.9984994674252254, "tokens_seen": 412221440 }, { "epoch": 0.12, "learning_rate": 0.00044195586760280845, "loss": 3.0161, "theoretical_loss": 3.998428713369126, "tokens_seen": 412286976 }, { "epoch": 0.12, "learning_rate": 0.00044194583751253763, "loss": 2.8992, "theoretical_loss": 3.9983579737075554, "tokens_seen": 412352512 }, { "epoch": 0.12, "learning_rate": 0.0004419358074222668, "loss": 2.9643, "theoretical_loss": 3.9982872484352976, "tokens_seen": 412418048 }, { "epoch": 0.12, "learning_rate": 0.000441925777331996, "loss": 3.0318, "theoretical_loss": 3.9982165375471403, "tokens_seen": 412483584 }, { "epoch": 0.13, "learning_rate": 0.0004419157472417252, "loss": 2.8583, "theoretical_loss": 3.998145841037873, "tokens_seen": 412549120 }, { "epoch": 0.13, "learning_rate": 0.0004419057171514544, "loss": 2.955, "theoretical_loss": 3.9980751589022887, "tokens_seen": 412614656 }, { "epoch": 0.13, "learning_rate": 0.00044189568706118354, "loss": 3.0281, "theoretical_loss": 3.9980044911351826, "tokens_seen": 412680192 }, { "epoch": 0.13, "learning_rate": 0.00044188565697091277, "loss": 2.9292, "theoretical_loss": 3.9979338377313525, "tokens_seen": 412745728 }, { "epoch": 0.13, "learning_rate": 0.0004418756268806419, "loss": 2.8697, "theoretical_loss": 3.997863198685599, "tokens_seen": 412811264 }, { "epoch": 0.13, "objective/train/docs_used": 245807, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.571214437484741, "objective/train/theoretical_loss": 3.997792573992726, "objective/train/tokens_used": 433336800, "theoretical_loss": 3.997792573992726, "tokens_seen": 412876800 }, { "epoch": 0.13, "learning_rate": 0.00044186559679037113, "loss": 2.7269, "theoretical_loss": 3.997792573992726, "tokens_seen": 412876800 }, { "epoch": 0.13, "learning_rate": 0.0004418555667001003, "loss": 3.0854, "theoretical_loss": 3.997721963647539, "tokens_seen": 412942336 }, { "epoch": 0.13, "learning_rate": 0.0004418455366098295, "loss": 3.036, "theoretical_loss": 3.997651367644848, "tokens_seen": 413007872 }, { "epoch": 0.13, "learning_rate": 0.0004418355065195587, "loss": 3.1661, "theoretical_loss": 3.997580785979464, "tokens_seen": 413073408 }, { "epoch": 0.13, "learning_rate": 0.0004418254764292879, "loss": 3.2061, "theoretical_loss": 3.9975102186462004, "tokens_seen": 413138944 }, { "epoch": 0.13, "learning_rate": 0.00044181544633901704, "loss": 2.8441, "theoretical_loss": 3.997439665639875, "tokens_seen": 413204480 }, { "epoch": 0.13, "learning_rate": 0.0004418054162487463, "loss": 2.9107, "theoretical_loss": 3.9973691269553076, "tokens_seen": 413270016 }, { "epoch": 0.13, "learning_rate": 0.0004417953861584754, "loss": 3.2179, "theoretical_loss": 3.9972986025873203, "tokens_seen": 413335552 }, { "epoch": 0.13, "learning_rate": 0.00044178535606820464, "loss": 2.8222, "theoretical_loss": 3.9972280925307384, "tokens_seen": 413401088 }, { "epoch": 0.13, "learning_rate": 0.0004417753259779338, "loss": 2.9046, "theoretical_loss": 3.9971575967803887, "tokens_seen": 413466624 }, { "epoch": 0.13, "learning_rate": 0.000441765295887663, "loss": 2.9597, "theoretical_loss": 3.9970871153311025, "tokens_seen": 413532160 }, { "epoch": 0.13, "learning_rate": 0.0004417552657973922, "loss": 2.7283, "theoretical_loss": 3.9970166481777127, "tokens_seen": 413597696 }, { "epoch": 0.13, "learning_rate": 0.00044174523570712136, "loss": 2.9719, "theoretical_loss": 3.996946195315055, "tokens_seen": 413663232 }, { "epoch": 0.13, "learning_rate": 0.00044173520561685054, "loss": 3.0133, "theoretical_loss": 3.996875756737968, "tokens_seen": 413728768 }, { "epoch": 0.13, "learning_rate": 0.0004417251755265798, "loss": 2.9793, "theoretical_loss": 3.996805332441293, "tokens_seen": 413794304 }, { "epoch": 0.13, "learning_rate": 0.0004417151454363089, "loss": 2.6928, "theoretical_loss": 3.996734922419873, "tokens_seen": 413859840 }, { "epoch": 0.13, "learning_rate": 0.00044170511534603814, "loss": 2.9173, "theoretical_loss": 3.996664526668555, "tokens_seen": 413925376 }, { "epoch": 0.13, "learning_rate": 0.00044169508525576727, "loss": 3.0386, "theoretical_loss": 3.9965941451821894, "tokens_seen": 413990912 }, { "epoch": 0.13, "learning_rate": 0.0004416850551654965, "loss": 2.9614, "theoretical_loss": 3.996523777955626, "tokens_seen": 414056448 }, { "epoch": 0.13, "learning_rate": 0.0004416750250752257, "loss": 3.1306, "theoretical_loss": 3.996453424983721, "tokens_seen": 414121984 }, { "epoch": 0.13, "learning_rate": 0.00044166499498495487, "loss": 2.9258, "theoretical_loss": 3.9963830862613303, "tokens_seen": 414187520 }, { "epoch": 0.13, "learning_rate": 0.00044165496489468405, "loss": 3.0298, "theoretical_loss": 3.9963127617833143, "tokens_seen": 414253056 }, { "epoch": 0.13, "learning_rate": 0.0004416449348044133, "loss": 2.9546, "theoretical_loss": 3.9962424515445356, "tokens_seen": 414318592 }, { "epoch": 0.13, "learning_rate": 0.0004416349047141424, "loss": 2.87, "theoretical_loss": 3.99617215553986, "tokens_seen": 414384128 }, { "epoch": 0.13, "learning_rate": 0.00044162487462387164, "loss": 3.0131, "theoretical_loss": 3.9961018737641547, "tokens_seen": 414449664 }, { "epoch": 0.13, "objective/train/docs_used": 246540, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8978941440582275, "objective/train/theoretical_loss": 3.99603160621229, "objective/train/tokens_used": 434975200, "theoretical_loss": 3.99603160621229, "tokens_seen": 414515200 }, { "epoch": 0.13, "learning_rate": 0.00044161484453360077, "loss": 2.9393, "theoretical_loss": 3.99603160621229, "tokens_seen": 414515200 }, { "epoch": 0.13, "learning_rate": 0.00044160481444333, "loss": 2.9509, "theoretical_loss": 3.9959613528791396, "tokens_seen": 414580736 }, { "epoch": 0.13, "learning_rate": 0.0004415947843530592, "loss": 2.7901, "theoretical_loss": 3.995891113759579, "tokens_seen": 414646272 }, { "epoch": 0.13, "learning_rate": 0.00044158475426278837, "loss": 2.9503, "theoretical_loss": 3.9958208888484865, "tokens_seen": 414711808 }, { "epoch": 0.13, "learning_rate": 0.00044157472417251755, "loss": 3.1345, "theoretical_loss": 3.9957506781407446, "tokens_seen": 414777344 }, { "epoch": 0.13, "learning_rate": 0.00044156469408224673, "loss": 2.7583, "theoretical_loss": 3.9956804816312355, "tokens_seen": 414842880 }, { "epoch": 0.13, "learning_rate": 0.0004415546639919759, "loss": 2.9844, "theoretical_loss": 3.995610299314846, "tokens_seen": 414908416 }, { "epoch": 0.13, "learning_rate": 0.00044154463390170515, "loss": 2.9255, "theoretical_loss": 3.9955401311864662, "tokens_seen": 414973952 }, { "epoch": 0.13, "learning_rate": 0.0004415346038114343, "loss": 3.0467, "theoretical_loss": 3.9954699772409867, "tokens_seen": 415039488 }, { "epoch": 0.13, "learning_rate": 0.0004415245737211635, "loss": 3.0262, "theoretical_loss": 3.995399837473302, "tokens_seen": 415105024 }, { "epoch": 0.13, "learning_rate": 0.00044151454363089264, "loss": 2.9524, "theoretical_loss": 3.9953297118783104, "tokens_seen": 415170560 }, { "epoch": 0.13, "learning_rate": 0.00044150451354062187, "loss": 3.0535, "theoretical_loss": 3.99525960045091, "tokens_seen": 415236096 }, { "epoch": 0.13, "learning_rate": 0.00044149448345035105, "loss": 2.8636, "theoretical_loss": 3.9951895031860043, "tokens_seen": 415301632 }, { "epoch": 0.13, "learning_rate": 0.00044148445336008023, "loss": 2.6274, "theoretical_loss": 3.9951194200784976, "tokens_seen": 415367168 }, { "epoch": 0.13, "learning_rate": 0.0004414744232698094, "loss": 3.1258, "theoretical_loss": 3.9950493511232974, "tokens_seen": 415432704 }, { "epoch": 0.13, "learning_rate": 0.00044146439317953865, "loss": 3.0572, "theoretical_loss": 3.9949792963153143, "tokens_seen": 415498240 }, { "epoch": 0.13, "learning_rate": 0.0004414543630892678, "loss": 3.019, "theoretical_loss": 3.9949092556494614, "tokens_seen": 415563776 }, { "epoch": 0.13, "learning_rate": 0.000441444332998997, "loss": 2.9243, "theoretical_loss": 3.994839229120654, "tokens_seen": 415629312 }, { "epoch": 0.13, "learning_rate": 0.0004414343029087262, "loss": 3.0326, "theoretical_loss": 3.99476921672381, "tokens_seen": 415694848 }, { "epoch": 0.13, "learning_rate": 0.0004414242728184554, "loss": 3.0729, "theoretical_loss": 3.99469921845385, "tokens_seen": 415760384 }, { "epoch": 0.13, "learning_rate": 0.0004414142427281846, "loss": 2.4204, "theoretical_loss": 3.994629234305698, "tokens_seen": 415825920 }, { "epoch": 0.13, "learning_rate": 0.00044140421263791374, "loss": 2.8157, "theoretical_loss": 3.9945592642742795, "tokens_seen": 415891456 }, { "epoch": 0.13, "learning_rate": 0.00044139418254764297, "loss": 3.0637, "theoretical_loss": 3.994489308354523, "tokens_seen": 415956992 }, { "epoch": 0.13, "learning_rate": 0.0004413841524573721, "loss": 3.0211, "theoretical_loss": 3.9944193665413605, "tokens_seen": 416022528 }, { "epoch": 0.13, "learning_rate": 0.00044137412236710133, "loss": 2.8919, "theoretical_loss": 3.9943494388297256, "tokens_seen": 416088064 }, { "epoch": 0.13, "objective/train/docs_used": 247824, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.898556709289551, "objective/train/theoretical_loss": 3.994279525214554, "objective/train/tokens_used": 436613600, "theoretical_loss": 3.994279525214554, "tokens_seen": 416153600 }, { "epoch": 0.13, "learning_rate": 0.0004413640922768305, "loss": 3.0149, "theoretical_loss": 3.994279525214554, "tokens_seen": 416153600 }, { "epoch": 0.13, "learning_rate": 0.0004413540621865597, "loss": 2.8584, "theoretical_loss": 3.9942096256907864, "tokens_seen": 416219136 }, { "epoch": 0.13, "learning_rate": 0.0004413440320962889, "loss": 3.0482, "theoretical_loss": 3.994139740253363, "tokens_seen": 416284672 }, { "epoch": 0.13, "learning_rate": 0.0004413340020060181, "loss": 2.8312, "theoretical_loss": 3.9940698688972285, "tokens_seen": 416350208 }, { "epoch": 0.13, "learning_rate": 0.00044132397191574724, "loss": 2.9454, "theoretical_loss": 3.99400001161733, "tokens_seen": 416415744 }, { "epoch": 0.13, "learning_rate": 0.0004413139418254765, "loss": 3.0826, "theoretical_loss": 3.9939301684086175, "tokens_seen": 416481280 }, { "epoch": 0.13, "learning_rate": 0.0004413039117352056, "loss": 2.4992, "theoretical_loss": 3.9938603392660426, "tokens_seen": 416546816 }, { "epoch": 0.13, "learning_rate": 0.00044129388164493484, "loss": 2.8932, "theoretical_loss": 3.9937905241845595, "tokens_seen": 416612352 }, { "epoch": 0.13, "learning_rate": 0.000441283851554664, "loss": 2.8243, "theoretical_loss": 3.993720723159127, "tokens_seen": 416677888 }, { "epoch": 0.13, "learning_rate": 0.0004412738214643932, "loss": 2.9899, "theoretical_loss": 3.9936509361847037, "tokens_seen": 416743424 }, { "epoch": 0.13, "learning_rate": 0.0004412637913741224, "loss": 2.5832, "theoretical_loss": 3.9935811632562532, "tokens_seen": 416808960 }, { "epoch": 0.13, "learning_rate": 0.00044125376128385156, "loss": 2.8921, "theoretical_loss": 3.99351140436874, "tokens_seen": 416874496 }, { "epoch": 0.13, "learning_rate": 0.00044124373119358074, "loss": 2.8575, "theoretical_loss": 3.9934416595171323, "tokens_seen": 416940032 }, { "epoch": 0.13, "learning_rate": 0.00044123370110331, "loss": 2.92, "theoretical_loss": 3.9933719286963996, "tokens_seen": 417005568 }, { "epoch": 0.13, "learning_rate": 0.0004412236710130391, "loss": 3.0291, "theoretical_loss": 3.993302211901516, "tokens_seen": 417071104 }, { "epoch": 0.13, "learning_rate": 0.00044121364092276834, "loss": 3.0544, "theoretical_loss": 3.9932325091274565, "tokens_seen": 417136640 }, { "epoch": 0.13, "learning_rate": 0.00044120361083249747, "loss": 2.965, "theoretical_loss": 3.993162820369199, "tokens_seen": 417202176 }, { "epoch": 0.13, "learning_rate": 0.0004411935807422267, "loss": 3.0694, "theoretical_loss": 3.9930931456217245, "tokens_seen": 417267712 }, { "epoch": 0.13, "learning_rate": 0.0004411835506519559, "loss": 2.9183, "theoretical_loss": 3.993023484880016, "tokens_seen": 417333248 }, { "epoch": 0.13, "learning_rate": 0.00044117352056168507, "loss": 2.9882, "theoretical_loss": 3.9929538381390595, "tokens_seen": 417398784 }, { "epoch": 0.13, "learning_rate": 0.00044116349047141425, "loss": 2.7475, "theoretical_loss": 3.9928842053938443, "tokens_seen": 417464320 }, { "epoch": 0.13, "learning_rate": 0.0004411534603811435, "loss": 3.0143, "theoretical_loss": 3.9928145866393594, "tokens_seen": 417529856 }, { "epoch": 0.13, "learning_rate": 0.0004411434302908726, "loss": 3.0615, "theoretical_loss": 3.9927449818706005, "tokens_seen": 417595392 }, { "epoch": 0.13, "learning_rate": 0.00044113340020060184, "loss": 3.0332, "theoretical_loss": 3.992675391082563, "tokens_seen": 417660928 }, { "epoch": 0.13, "learning_rate": 0.00044112337011033097, "loss": 2.7898, "theoretical_loss": 3.9926058142702456, "tokens_seen": 417726464 }, { "epoch": 0.13, "objective/train/docs_used": 248559, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1589126586914062, "objective/train/theoretical_loss": 3.9925362514286498, "objective/train/tokens_used": 438252000, "theoretical_loss": 3.9925362514286498, "tokens_seen": 417792000 }, { "epoch": 0.13, "learning_rate": 0.0004411133400200602, "loss": 2.8932, "theoretical_loss": 3.9925362514286498, "tokens_seen": 417792000 }, { "epoch": 0.13, "learning_rate": 0.0004411033099297894, "loss": 3.1841, "theoretical_loss": 3.992466702552779, "tokens_seen": 417857536 }, { "epoch": 0.13, "learning_rate": 0.00044109327983951857, "loss": 2.9222, "theoretical_loss": 3.99239716763764, "tokens_seen": 417923072 }, { "epoch": 0.13, "learning_rate": 0.00044108324974924775, "loss": 3.1899, "theoretical_loss": 3.9923276466782425, "tokens_seen": 417988608 }, { "epoch": 0.13, "learning_rate": 0.00044107321965897693, "loss": 2.9757, "theoretical_loss": 3.9922581396695973, "tokens_seen": 418054144 }, { "epoch": 0.13, "learning_rate": 0.0004410631895687061, "loss": 3.003, "theoretical_loss": 3.992188646606719, "tokens_seen": 418119680 }, { "epoch": 0.13, "learning_rate": 0.00044105315947843535, "loss": 2.766, "theoretical_loss": 3.992119167484624, "tokens_seen": 418185216 }, { "epoch": 0.13, "learning_rate": 0.0004410431293881645, "loss": 2.8755, "theoretical_loss": 3.992049702298332, "tokens_seen": 418250752 }, { "epoch": 0.13, "learning_rate": 0.0004410330992978937, "loss": 2.818, "theoretical_loss": 3.9919802510428646, "tokens_seen": 418316288 }, { "epoch": 0.13, "learning_rate": 0.00044102306920762284, "loss": 2.9323, "theoretical_loss": 3.991910813713246, "tokens_seen": 418381824 }, { "epoch": 0.13, "learning_rate": 0.00044101303911735207, "loss": 2.9617, "theoretical_loss": 3.991841390304504, "tokens_seen": 418447360 }, { "epoch": 0.13, "learning_rate": 0.00044100300902708125, "loss": 3.0045, "theoretical_loss": 3.991771980811667, "tokens_seen": 418512896 }, { "epoch": 0.13, "learning_rate": 0.00044099297893681043, "loss": 2.9303, "theoretical_loss": 3.991702585229768, "tokens_seen": 418578432 }, { "epoch": 0.13, "learning_rate": 0.0004409829488465396, "loss": 3.0384, "theoretical_loss": 3.9916332035538415, "tokens_seen": 418643968 }, { "epoch": 0.13, "learning_rate": 0.00044097291875626885, "loss": 2.9644, "theoretical_loss": 3.9915638357789245, "tokens_seen": 418709504 }, { "epoch": 0.13, "learning_rate": 0.000440962888665998, "loss": 2.9537, "theoretical_loss": 3.991494481900056, "tokens_seen": 418775040 }, { "epoch": 0.13, "learning_rate": 0.0004409528585757272, "loss": 3.1591, "theoretical_loss": 3.9914251419122797, "tokens_seen": 418840576 }, { "epoch": 0.13, "learning_rate": 0.00044094282848545634, "loss": 2.9929, "theoretical_loss": 3.99135581581064, "tokens_seen": 418906112 }, { "epoch": 0.13, "learning_rate": 0.0004409327983951856, "loss": 2.9587, "theoretical_loss": 3.991286503590184, "tokens_seen": 418971648 }, { "epoch": 0.13, "learning_rate": 0.00044092276830491476, "loss": 2.9491, "theoretical_loss": 3.9912172052459614, "tokens_seen": 419037184 }, { "epoch": 0.13, "learning_rate": 0.00044091273821464394, "loss": 3.0031, "theoretical_loss": 3.9911479207730247, "tokens_seen": 419102720 }, { "epoch": 0.13, "learning_rate": 0.0004409027081243731, "loss": 3.0363, "theoretical_loss": 3.991078650166429, "tokens_seen": 419168256 }, { "epoch": 0.13, "learning_rate": 0.0004408926780341023, "loss": 2.9847, "theoretical_loss": 3.9910093934212325, "tokens_seen": 419233792 }, { "epoch": 0.13, "learning_rate": 0.0004408826479438315, "loss": 2.8094, "theoretical_loss": 3.990940150532494, "tokens_seen": 419299328 }, { "epoch": 0.13, "learning_rate": 0.0004408726178535607, "loss": 2.831, "theoretical_loss": 3.990870921495277, "tokens_seen": 419364864 }, { "epoch": 0.13, "objective/train/docs_used": 249785, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2360668182373047, "objective/train/theoretical_loss": 3.990801706304647, "objective/train/tokens_used": 439890400, "theoretical_loss": 3.990801706304647, "tokens_seen": 419430400 }, { "epoch": 0.13, "learning_rate": 0.00044086258776328984, "loss": 3.028, "theoretical_loss": 3.990801706304647, "tokens_seen": 419430400 }, { "epoch": 0.13, "learning_rate": 0.0004408525576730191, "loss": 3.0799, "theoretical_loss": 3.99073250495567, "tokens_seen": 419495936 }, { "epoch": 0.13, "learning_rate": 0.0004408425275827482, "loss": 3.1414, "theoretical_loss": 3.990663317443418, "tokens_seen": 419561472 }, { "epoch": 0.13, "learning_rate": 0.00044083249749247744, "loss": 2.9101, "theoretical_loss": 3.990594143762962, "tokens_seen": 419627008 }, { "epoch": 0.13, "learning_rate": 0.0004408224674022066, "loss": 2.9956, "theoretical_loss": 3.9905249839093786, "tokens_seen": 419692544 }, { "epoch": 0.13, "learning_rate": 0.0004408124373119358, "loss": 2.7741, "theoretical_loss": 3.9904558378777453, "tokens_seen": 419758080 }, { "epoch": 0.13, "learning_rate": 0.000440802407221665, "loss": 2.8678, "theoretical_loss": 3.990386705663142, "tokens_seen": 419823616 }, { "epoch": 0.13, "learning_rate": 0.0004407923771313942, "loss": 2.7025, "theoretical_loss": 3.9903175872606518, "tokens_seen": 419889152 }, { "epoch": 0.13, "learning_rate": 0.00044078234704112335, "loss": 2.9144, "theoretical_loss": 3.990248482665359, "tokens_seen": 419954688 }, { "epoch": 0.13, "learning_rate": 0.0004407723169508526, "loss": 2.9458, "theoretical_loss": 3.9901793918723527, "tokens_seen": 420020224 }, { "epoch": 0.13, "learning_rate": 0.0004407622868605817, "loss": 2.8895, "theoretical_loss": 3.9901103148767225, "tokens_seen": 420085760 }, { "epoch": 0.13, "learning_rate": 0.00044075225677031094, "loss": 2.9166, "theoretical_loss": 3.9900412516735617, "tokens_seen": 420151296 }, { "epoch": 0.13, "learning_rate": 0.0004407422266800401, "loss": 3.0461, "theoretical_loss": 3.989972202257966, "tokens_seen": 420216832 }, { "epoch": 0.13, "learning_rate": 0.0004407321965897693, "loss": 3.2388, "theoretical_loss": 3.989903166625032, "tokens_seen": 420282368 }, { "epoch": 0.13, "learning_rate": 0.0004407221664994985, "loss": 2.6707, "theoretical_loss": 3.989834144769861, "tokens_seen": 420347904 }, { "epoch": 0.13, "learning_rate": 0.00044071213640922767, "loss": 2.9935, "theoretical_loss": 3.9897651366875557, "tokens_seen": 420413440 }, { "epoch": 0.13, "learning_rate": 0.00044070210631895685, "loss": 3.0185, "theoretical_loss": 3.989696142373221, "tokens_seen": 420478976 }, { "epoch": 0.13, "learning_rate": 0.0004406920762286861, "loss": 2.7856, "theoretical_loss": 3.9896271618219656, "tokens_seen": 420544512 }, { "epoch": 0.13, "learning_rate": 0.00044068204613841527, "loss": 2.7666, "theoretical_loss": 3.9895581950288994, "tokens_seen": 420610048 }, { "epoch": 0.13, "learning_rate": 0.00044067201604814445, "loss": 3.0835, "theoretical_loss": 3.9894892419891352, "tokens_seen": 420675584 }, { "epoch": 0.13, "learning_rate": 0.0004406619859578737, "loss": 2.7458, "theoretical_loss": 3.989420302697789, "tokens_seen": 420741120 }, { "epoch": 0.13, "learning_rate": 0.0004406519558676028, "loss": 2.9788, "theoretical_loss": 3.989351377149978, "tokens_seen": 420806656 }, { "epoch": 0.13, "learning_rate": 0.00044064192577733204, "loss": 3.0151, "theoretical_loss": 3.989282465340823, "tokens_seen": 420872192 }, { "epoch": 0.13, "learning_rate": 0.00044063189568706117, "loss": 2.7167, "theoretical_loss": 3.9892135672654465, "tokens_seen": 420937728 }, { "epoch": 0.13, "learning_rate": 0.0004406218655967904, "loss": 2.7923, "theoretical_loss": 3.9891446829189734, "tokens_seen": 421003264 }, { "epoch": 0.13, "objective/train/docs_used": 250392, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.840327262878418, "objective/train/theoretical_loss": 3.989075812296533, "objective/train/tokens_used": 441528800, "theoretical_loss": 3.989075812296533, "tokens_seen": 421068800 }, { "epoch": 0.13, "learning_rate": 0.0004406118355065196, "loss": 2.9502, "theoretical_loss": 3.989075812296533, "tokens_seen": 421068800 }, { "epoch": 0.13, "learning_rate": 0.00044060180541624877, "loss": 2.9474, "theoretical_loss": 3.989006955393254, "tokens_seen": 421134336 }, { "epoch": 0.13, "learning_rate": 0.00044059177532597795, "loss": 2.8847, "theoretical_loss": 3.98893811220427, "tokens_seen": 421199872 }, { "epoch": 0.13, "learning_rate": 0.00044058174523570713, "loss": 3.1015, "theoretical_loss": 3.9888692827247167, "tokens_seen": 421265408 }, { "epoch": 0.13, "learning_rate": 0.0004405717151454363, "loss": 2.8387, "theoretical_loss": 3.9888004669497317, "tokens_seen": 421330944 }, { "epoch": 0.13, "learning_rate": 0.00044056168505516555, "loss": 2.9615, "theoretical_loss": 3.988731664874454, "tokens_seen": 421396480 }, { "epoch": 0.13, "learning_rate": 0.0004405516549648947, "loss": 2.9015, "theoretical_loss": 3.9886628764940277, "tokens_seen": 421462016 }, { "epoch": 0.13, "learning_rate": 0.0004405416248746239, "loss": 2.8663, "theoretical_loss": 3.988594101803598, "tokens_seen": 421527552 }, { "epoch": 0.13, "learning_rate": 0.00044053159478435304, "loss": 2.9738, "theoretical_loss": 3.9885253407983114, "tokens_seen": 421593088 }, { "epoch": 0.13, "learning_rate": 0.00044052156469408227, "loss": 2.8473, "theoretical_loss": 3.9884565934733196, "tokens_seen": 421658624 }, { "epoch": 0.13, "learning_rate": 0.00044051153460381145, "loss": 3.1179, "theoretical_loss": 3.9883878598237743, "tokens_seen": 421724160 }, { "epoch": 0.13, "learning_rate": 0.00044050150451354063, "loss": 2.9468, "theoretical_loss": 3.9883191398448306, "tokens_seen": 421789696 }, { "epoch": 0.13, "learning_rate": 0.0004404914744232698, "loss": 3.0772, "theoretical_loss": 3.9882504335316464, "tokens_seen": 421855232 }, { "epoch": 0.13, "learning_rate": 0.00044048144433299905, "loss": 2.9365, "theoretical_loss": 3.988181740879382, "tokens_seen": 421920768 }, { "epoch": 0.13, "learning_rate": 0.0004404714142427282, "loss": 2.6722, "theoretical_loss": 3.9881130618831984, "tokens_seen": 421986304 }, { "epoch": 0.13, "learning_rate": 0.0004404613841524574, "loss": 3.0752, "theoretical_loss": 3.988044396538263, "tokens_seen": 422051840 }, { "epoch": 0.13, "learning_rate": 0.00044045135406218654, "loss": 2.6501, "theoretical_loss": 3.9879757448397406, "tokens_seen": 422117376 }, { "epoch": 0.13, "learning_rate": 0.0004404413239719158, "loss": 2.9492, "theoretical_loss": 3.9879071067828034, "tokens_seen": 422182912 }, { "epoch": 0.13, "learning_rate": 0.00044043129388164496, "loss": 2.8998, "theoretical_loss": 3.9878384823626227, "tokens_seen": 422248448 }, { "epoch": 0.13, "learning_rate": 0.00044042126379137414, "loss": 2.898, "theoretical_loss": 3.987769871574373, "tokens_seen": 422313984 }, { "epoch": 0.13, "learning_rate": 0.0004404112337011033, "loss": 3.1466, "theoretical_loss": 3.9877012744132316, "tokens_seen": 422379520 }, { "epoch": 0.13, "learning_rate": 0.0004404012036108325, "loss": 2.8585, "theoretical_loss": 3.987632690874379, "tokens_seen": 422445056 }, { "epoch": 0.13, "learning_rate": 0.0004403911735205617, "loss": 3.0978, "theoretical_loss": 3.987564120952997, "tokens_seen": 422510592 }, { "epoch": 0.13, "learning_rate": 0.0004403811434302909, "loss": 2.8917, "theoretical_loss": 3.98749556464427, "tokens_seen": 422576128 }, { "epoch": 0.13, "learning_rate": 0.00044037111334002004, "loss": 2.8132, "theoretical_loss": 3.987427021943385, "tokens_seen": 422641664 }, { "epoch": 0.13, "objective/train/docs_used": 251199, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.547407627105713, "objective/train/theoretical_loss": 3.987358492845532, "objective/train/tokens_used": 443167200, "theoretical_loss": 3.987358492845532, "tokens_seen": 422707200 }, { "epoch": 0.13, "learning_rate": 0.0004403610832497493, "loss": 2.9726, "theoretical_loss": 3.987358492845532, "tokens_seen": 422707200 }, { "epoch": 0.13, "learning_rate": 0.0004403510531594784, "loss": 3.0115, "theoretical_loss": 3.9872899773459025, "tokens_seen": 422772736 }, { "epoch": 0.13, "learning_rate": 0.00044034102306920764, "loss": 2.8462, "theoretical_loss": 3.987221475439691, "tokens_seen": 422838272 }, { "epoch": 0.13, "learning_rate": 0.0004403309929789368, "loss": 2.8494, "theoretical_loss": 3.9871529871220943, "tokens_seen": 422903808 }, { "epoch": 0.13, "learning_rate": 0.000440320962888666, "loss": 2.5373, "theoretical_loss": 3.9870845123883116, "tokens_seen": 422969344 }, { "epoch": 0.13, "learning_rate": 0.0004403109327983952, "loss": 3.0218, "theoretical_loss": 3.987016051233546, "tokens_seen": 423034880 }, { "epoch": 0.13, "learning_rate": 0.0004403009027081244, "loss": 2.9801, "theoretical_loss": 3.9869476036529994, "tokens_seen": 423100416 }, { "epoch": 0.13, "learning_rate": 0.00044029087261785355, "loss": 2.9104, "theoretical_loss": 3.98687916964188, "tokens_seen": 423165952 }, { "epoch": 0.13, "learning_rate": 0.0004402808425275828, "loss": 2.9944, "theoretical_loss": 3.986810749195396, "tokens_seen": 423231488 }, { "epoch": 0.13, "learning_rate": 0.0004402708124373119, "loss": 2.9082, "theoretical_loss": 3.9867423423087596, "tokens_seen": 423297024 }, { "epoch": 0.13, "learning_rate": 0.00044026078234704114, "loss": 2.9377, "theoretical_loss": 3.9866739489771845, "tokens_seen": 423362560 }, { "epoch": 0.13, "learning_rate": 0.0004402507522567703, "loss": 3.0668, "theoretical_loss": 3.986605569195887, "tokens_seen": 423428096 }, { "epoch": 0.13, "learning_rate": 0.0004402407221664995, "loss": 3.0737, "theoretical_loss": 3.986537202960085, "tokens_seen": 423493632 }, { "epoch": 0.13, "learning_rate": 0.0004402306920762287, "loss": 3.0872, "theoretical_loss": 3.986468850265001, "tokens_seen": 423559168 }, { "epoch": 0.13, "learning_rate": 0.00044022066198595787, "loss": 3.1557, "theoretical_loss": 3.986400511105858, "tokens_seen": 423624704 }, { "epoch": 0.13, "learning_rate": 0.00044021063189568705, "loss": 3.0565, "theoretical_loss": 3.9863321854778815, "tokens_seen": 423690240 }, { "epoch": 0.13, "learning_rate": 0.0004402006018054163, "loss": 3.1438, "theoretical_loss": 3.986263873376301, "tokens_seen": 423755776 }, { "epoch": 0.13, "learning_rate": 0.0004401905717151454, "loss": 2.8585, "theoretical_loss": 3.986195574796347, "tokens_seen": 423821312 }, { "epoch": 0.13, "learning_rate": 0.00044018054162487465, "loss": 2.5956, "theoretical_loss": 3.9861272897332527, "tokens_seen": 423886848 }, { "epoch": 0.13, "learning_rate": 0.00044017051153460383, "loss": 2.902, "theoretical_loss": 3.986059018182254, "tokens_seen": 423952384 }, { "epoch": 0.13, "learning_rate": 0.000440160481444333, "loss": 2.987, "theoretical_loss": 3.9859907601385887, "tokens_seen": 424017920 }, { "epoch": 0.13, "learning_rate": 0.0004401504513540622, "loss": 2.6594, "theoretical_loss": 3.9859225155974976, "tokens_seen": 424083456 }, { "epoch": 0.13, "learning_rate": 0.00044014042126379137, "loss": 2.9757, "theoretical_loss": 3.985854284554223, "tokens_seen": 424148992 }, { "epoch": 0.13, "learning_rate": 0.00044013039117352055, "loss": 3.1121, "theoretical_loss": 3.9857860670040113, "tokens_seen": 424214528 }, { "epoch": 0.13, "learning_rate": 0.0004401203610832498, "loss": 3.074, "theoretical_loss": 3.98571786294211, "tokens_seen": 424280064 }, { "epoch": 0.13, "objective/train/docs_used": 252247, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.094590187072754, "objective/train/theoretical_loss": 3.9856496723637687, "objective/train/tokens_used": 444805600, "theoretical_loss": 3.9856496723637687, "tokens_seen": 424345600 }, { "epoch": 0.13, "learning_rate": 0.0004401103309929789, "loss": 3.0113, "theoretical_loss": 3.9856496723637687, "tokens_seen": 424345600 }, { "epoch": 0.13, "learning_rate": 0.00044010030090270815, "loss": 3.0729, "theoretical_loss": 3.985581495264241, "tokens_seen": 424411136 }, { "epoch": 0.13, "learning_rate": 0.0004400902708124373, "loss": 2.815, "theoretical_loss": 3.9855133316387805, "tokens_seen": 424476672 }, { "epoch": 0.13, "learning_rate": 0.0004400802407221665, "loss": 3.0778, "theoretical_loss": 3.9854451814826457, "tokens_seen": 424542208 }, { "epoch": 0.13, "learning_rate": 0.0004400702106318957, "loss": 2.8332, "theoretical_loss": 3.985377044791096, "tokens_seen": 424607744 }, { "epoch": 0.13, "learning_rate": 0.0004400601805416249, "loss": 2.9363, "theoretical_loss": 3.985308921559394, "tokens_seen": 424673280 }, { "epoch": 0.13, "learning_rate": 0.00044005015045135406, "loss": 2.8434, "theoretical_loss": 3.9852408117828038, "tokens_seen": 424738816 }, { "epoch": 0.13, "learning_rate": 0.00044004012036108324, "loss": 2.9161, "theoretical_loss": 3.9851727154565926, "tokens_seen": 424804352 }, { "epoch": 0.13, "learning_rate": 0.0004400300902708124, "loss": 2.7819, "theoretical_loss": 3.98510463257603, "tokens_seen": 424869888 }, { "epoch": 0.13, "learning_rate": 0.00044002006018054165, "loss": 3.1194, "theoretical_loss": 3.9850365631363873, "tokens_seen": 424935424 }, { "epoch": 0.13, "learning_rate": 0.0004400100300902708, "loss": 2.7506, "theoretical_loss": 3.9849685071329386, "tokens_seen": 425000960 }, { "epoch": 0.13, "learning_rate": 0.00044, "loss": 3.132, "theoretical_loss": 3.984900464560961, "tokens_seen": 425066496 }, { "epoch": 0.13, "learning_rate": 0.0004399899699097292, "loss": 2.8485, "theoretical_loss": 3.984832435415733, "tokens_seen": 425132032 }, { "epoch": 0.13, "learning_rate": 0.0004399799398194584, "loss": 2.9311, "theoretical_loss": 3.984764419692537, "tokens_seen": 425197568 }, { "epoch": 0.13, "learning_rate": 0.00043996990972918756, "loss": 2.9116, "theoretical_loss": 3.984696417386655, "tokens_seen": 425263104 }, { "epoch": 0.13, "learning_rate": 0.00043995987963891674, "loss": 2.7839, "theoretical_loss": 3.984628428493374, "tokens_seen": 425328640 }, { "epoch": 0.13, "learning_rate": 0.0004399498495486459, "loss": 2.6203, "theoretical_loss": 3.9845604530079832, "tokens_seen": 425394176 }, { "epoch": 0.13, "learning_rate": 0.00043993981945837516, "loss": 2.6998, "theoretical_loss": 3.984492490925772, "tokens_seen": 425459712 }, { "epoch": 0.13, "learning_rate": 0.00043992978936810434, "loss": 2.9726, "theoretical_loss": 3.984424542242034, "tokens_seen": 425525248 }, { "epoch": 0.13, "learning_rate": 0.0004399197592778335, "loss": 2.9377, "theoretical_loss": 3.9843566069520664, "tokens_seen": 425590784 }, { "epoch": 0.13, "learning_rate": 0.0004399097291875627, "loss": 2.6459, "theoretical_loss": 3.9842886850511654, "tokens_seen": 425656320 }, { "epoch": 0.13, "learning_rate": 0.0004398996990972919, "loss": 2.7193, "theoretical_loss": 3.984220776534632, "tokens_seen": 425721856 }, { "epoch": 0.13, "learning_rate": 0.0004398896690070211, "loss": 2.8158, "theoretical_loss": 3.984152881397769, "tokens_seen": 425787392 }, { "epoch": 0.13, "learning_rate": 0.00043987963891675024, "loss": 2.9572, "theoretical_loss": 3.9840849996358814, "tokens_seen": 425852928 }, { "epoch": 0.13, "learning_rate": 0.0004398696088264795, "loss": 2.6641, "theoretical_loss": 3.9840171312442765, "tokens_seen": 425918464 }, { "epoch": 0.13, "objective/train/docs_used": 253004, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.064711093902588, "objective/train/theoretical_loss": 3.9839492762182647, "objective/train/tokens_used": 446444000, "theoretical_loss": 3.9839492762182647, "tokens_seen": 425984000 }, { "epoch": 0.13, "learning_rate": 0.0004398595787362086, "loss": 2.9181, "theoretical_loss": 3.9839492762182647, "tokens_seen": 425984000 }, { "epoch": 0.13, "learning_rate": 0.00043984954864593784, "loss": 3.0364, "theoretical_loss": 3.9838814345531577, "tokens_seen": 426049536 }, { "epoch": 0.13, "learning_rate": 0.000439839518555667, "loss": 2.7298, "theoretical_loss": 3.9838136062442704, "tokens_seen": 426115072 }, { "epoch": 0.13, "learning_rate": 0.0004398294884653962, "loss": 2.9424, "theoretical_loss": 3.98374579128692, "tokens_seen": 426180608 }, { "epoch": 0.13, "learning_rate": 0.0004398194583751254, "loss": 2.9639, "theoretical_loss": 3.9836779896764254, "tokens_seen": 426246144 }, { "epoch": 0.13, "learning_rate": 0.0004398094282848546, "loss": 2.8545, "theoretical_loss": 3.9836102014081076, "tokens_seen": 426311680 }, { "epoch": 0.13, "learning_rate": 0.00043979939819458375, "loss": 2.826, "theoretical_loss": 3.983542426477292, "tokens_seen": 426377216 }, { "epoch": 0.13, "learning_rate": 0.000439789368104313, "loss": 2.8631, "theoretical_loss": 3.9834746648793042, "tokens_seen": 426442752 }, { "epoch": 0.13, "learning_rate": 0.0004397793380140421, "loss": 2.925, "theoretical_loss": 3.983406916609473, "tokens_seen": 426508288 }, { "epoch": 0.13, "learning_rate": 0.00043976930792377134, "loss": 3.0121, "theoretical_loss": 3.98333918166313, "tokens_seen": 426573824 }, { "epoch": 0.13, "learning_rate": 0.0004397592778335005, "loss": 2.9686, "theoretical_loss": 3.983271460035608, "tokens_seen": 426639360 }, { "epoch": 0.13, "learning_rate": 0.0004397492477432297, "loss": 2.628, "theoretical_loss": 3.983203751722243, "tokens_seen": 426704896 }, { "epoch": 0.13, "learning_rate": 0.0004397392176529589, "loss": 2.8485, "theoretical_loss": 3.983136056718373, "tokens_seen": 426770432 }, { "epoch": 0.13, "learning_rate": 0.00043972918756268807, "loss": 2.8924, "theoretical_loss": 3.9830683750193394, "tokens_seen": 426835968 }, { "epoch": 0.13, "learning_rate": 0.00043971915747241725, "loss": 2.8836, "theoretical_loss": 3.9830007066204836, "tokens_seen": 426901504 }, { "epoch": 0.13, "learning_rate": 0.0004397091273821465, "loss": 2.893, "theoretical_loss": 3.9829330515171524, "tokens_seen": 426967040 }, { "epoch": 0.13, "learning_rate": 0.0004396990972918756, "loss": 2.8905, "theoretical_loss": 3.982865409704692, "tokens_seen": 427032576 }, { "epoch": 0.13, "learning_rate": 0.00043968906720160485, "loss": 2.9351, "theoretical_loss": 3.9827977811784523, "tokens_seen": 427098112 }, { "epoch": 0.13, "learning_rate": 0.00043967903711133403, "loss": 2.8235, "theoretical_loss": 3.9827301659337864, "tokens_seen": 427163648 }, { "epoch": 0.13, "learning_rate": 0.0004396690070210632, "loss": 3.0057, "theoretical_loss": 3.9826625639660485, "tokens_seen": 427229184 }, { "epoch": 0.13, "learning_rate": 0.0004396589769307924, "loss": 2.958, "theoretical_loss": 3.982594975270595, "tokens_seen": 427294720 }, { "epoch": 0.13, "learning_rate": 0.00043964894684052157, "loss": 2.8409, "theoretical_loss": 3.982527399842786, "tokens_seen": 427360256 }, { "epoch": 0.13, "learning_rate": 0.00043963891675025075, "loss": 2.7254, "theoretical_loss": 3.9824598376779825, "tokens_seen": 427425792 }, { "epoch": 0.13, "learning_rate": 0.00043962888665998, "loss": 2.9997, "theoretical_loss": 3.982392288771549, "tokens_seen": 427491328 }, { "epoch": 0.13, "learning_rate": 0.0004396188565697091, "loss": 2.7354, "theoretical_loss": 3.98232475311885, "tokens_seen": 427556864 }, { "epoch": 0.13, "objective/train/docs_used": 254260, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1842780113220215, "objective/train/theoretical_loss": 3.9822572307152564, "objective/train/tokens_used": 448082400, "theoretical_loss": 3.9822572307152564, "tokens_seen": 427622400 }, { "epoch": 0.13, "learning_rate": 0.00043960882647943835, "loss": 2.9841, "theoretical_loss": 3.9822572307152564, "tokens_seen": 427622400 }, { "epoch": 0.13, "learning_rate": 0.0004395987963891675, "loss": 3.3788, "theoretical_loss": 3.9821897215561375, "tokens_seen": 427687936 }, { "epoch": 0.13, "learning_rate": 0.0004395887662988967, "loss": 2.8013, "theoretical_loss": 3.982122225636867, "tokens_seen": 427753472 }, { "epoch": 0.13, "learning_rate": 0.0004395787362086259, "loss": 3.0004, "theoretical_loss": 3.98205474295282, "tokens_seen": 427819008 }, { "epoch": 0.13, "learning_rate": 0.0004395687061183551, "loss": 2.9367, "theoretical_loss": 3.9819872734993753, "tokens_seen": 427884544 }, { "epoch": 0.13, "learning_rate": 0.00043955867602808426, "loss": 2.971, "theoretical_loss": 3.9819198172719124, "tokens_seen": 427950080 }, { "epoch": 0.13, "learning_rate": 0.00043954864593781344, "loss": 2.8509, "theoretical_loss": 3.981852374265814, "tokens_seen": 428015616 }, { "epoch": 0.13, "learning_rate": 0.0004395386158475426, "loss": 2.9302, "theoretical_loss": 3.9817849444764644, "tokens_seen": 428081152 }, { "epoch": 0.13, "learning_rate": 0.00043952858575727185, "loss": 2.7305, "theoretical_loss": 3.981717527899252, "tokens_seen": 428146688 }, { "epoch": 0.13, "learning_rate": 0.000439518555667001, "loss": 2.7329, "theoretical_loss": 3.9816501245295646, "tokens_seen": 428212224 }, { "epoch": 0.13, "learning_rate": 0.0004395085255767302, "loss": 2.95, "theoretical_loss": 3.9815827343627954, "tokens_seen": 428277760 }, { "epoch": 0.13, "learning_rate": 0.0004394984954864594, "loss": 2.7096, "theoretical_loss": 3.981515357394338, "tokens_seen": 428343296 }, { "epoch": 0.13, "learning_rate": 0.0004394884653961886, "loss": 3.0307, "theoretical_loss": 3.9814479936195886, "tokens_seen": 428408832 }, { "epoch": 0.13, "learning_rate": 0.00043947843530591776, "loss": 2.6595, "theoretical_loss": 3.981380643033945, "tokens_seen": 428474368 }, { "epoch": 0.13, "learning_rate": 0.00043946840521564694, "loss": 2.6865, "theoretical_loss": 3.9813133056328103, "tokens_seen": 428539904 }, { "epoch": 0.13, "learning_rate": 0.0004394583751253761, "loss": 3.0187, "theoretical_loss": 3.981245981411586, "tokens_seen": 428605440 }, { "epoch": 0.13, "learning_rate": 0.00043944834503510536, "loss": 2.813, "theoretical_loss": 3.981178670365679, "tokens_seen": 428670976 }, { "epoch": 0.13, "learning_rate": 0.0004394383149448345, "loss": 2.8802, "theoretical_loss": 3.981111372490496, "tokens_seen": 428736512 }, { "epoch": 0.13, "learning_rate": 0.0004394282848545637, "loss": 2.9085, "theoretical_loss": 3.9810440877814486, "tokens_seen": 428802048 }, { "epoch": 0.13, "learning_rate": 0.00043941825476429285, "loss": 2.5354, "theoretical_loss": 3.9809768162339476, "tokens_seen": 428867584 }, { "epoch": 0.13, "learning_rate": 0.0004394082246740221, "loss": 2.8229, "theoretical_loss": 3.9809095578434093, "tokens_seen": 428933120 }, { "epoch": 0.13, "learning_rate": 0.00043939819458375126, "loss": 3.0868, "theoretical_loss": 3.9808423126052506, "tokens_seen": 428998656 }, { "epoch": 0.13, "learning_rate": 0.00043938816449348044, "loss": 2.8457, "theoretical_loss": 3.9807750805148903, "tokens_seen": 429064192 }, { "epoch": 0.13, "learning_rate": 0.0004393781344032096, "loss": 2.9069, "theoretical_loss": 3.9807078615677502, "tokens_seen": 429129728 }, { "epoch": 0.13, "learning_rate": 0.0004393681043129388, "loss": 3.0153, "theoretical_loss": 3.980640655759255, "tokens_seen": 429195264 }, { "epoch": 0.13, "objective/train/docs_used": 255066, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0367283821105957, "objective/train/theoretical_loss": 3.9805734630848306, "objective/train/tokens_used": 449720800, "theoretical_loss": 3.9805734630848306, "tokens_seen": 429260800 }, { "epoch": 0.13, "learning_rate": 0.000439358074222668, "loss": 2.8789, "theoretical_loss": 3.9805734630848306, "tokens_seen": 429260800 }, { "epoch": 0.13, "learning_rate": 0.0004393480441323972, "loss": 3.0386, "theoretical_loss": 3.9805062835399054, "tokens_seen": 429326336 }, { "epoch": 0.13, "learning_rate": 0.00043933801404212635, "loss": 2.7223, "theoretical_loss": 3.9804391171199103, "tokens_seen": 429391872 }, { "epoch": 0.13, "learning_rate": 0.0004393279839518556, "loss": 2.8458, "theoretical_loss": 3.9803719638202786, "tokens_seen": 429457408 }, { "epoch": 0.13, "learning_rate": 0.00043931795386158477, "loss": 2.8339, "theoretical_loss": 3.9803048236364456, "tokens_seen": 429522944 }, { "epoch": 0.13, "learning_rate": 0.00043930792377131395, "loss": 3.011, "theoretical_loss": 3.9802376965638495, "tokens_seen": 429588480 }, { "epoch": 0.13, "learning_rate": 0.00043929789368104313, "loss": 3.0028, "theoretical_loss": 3.98017058259793, "tokens_seen": 429654016 }, { "epoch": 0.13, "learning_rate": 0.0004392878635907723, "loss": 2.925, "theoretical_loss": 3.980103481734129, "tokens_seen": 429719552 }, { "epoch": 0.13, "learning_rate": 0.0004392778335005015, "loss": 2.9888, "theoretical_loss": 3.9800363939678918, "tokens_seen": 429785088 }, { "epoch": 0.13, "learning_rate": 0.0004392678034102307, "loss": 3.0124, "theoretical_loss": 3.9799693192946646, "tokens_seen": 429850624 }, { "epoch": 0.13, "learning_rate": 0.00043925777331995985, "loss": 2.8369, "theoretical_loss": 3.979902257709897, "tokens_seen": 429916160 }, { "epoch": 0.13, "learning_rate": 0.0004392477432296891, "loss": 3.0397, "theoretical_loss": 3.97983520920904, "tokens_seen": 429981696 }, { "epoch": 0.13, "learning_rate": 0.0004392377131394182, "loss": 2.8115, "theoretical_loss": 3.979768173787548, "tokens_seen": 430047232 }, { "epoch": 0.13, "learning_rate": 0.00043922768304914745, "loss": 3.0784, "theoretical_loss": 3.9797011514408758, "tokens_seen": 430112768 }, { "epoch": 0.13, "learning_rate": 0.00043921765295887663, "loss": 2.6783, "theoretical_loss": 3.9796341421644827, "tokens_seen": 430178304 }, { "epoch": 0.13, "learning_rate": 0.0004392076228686058, "loss": 2.7325, "theoretical_loss": 3.979567145953829, "tokens_seen": 430243840 }, { "epoch": 0.13, "learning_rate": 0.000439197592778335, "loss": 2.8376, "theoretical_loss": 3.979500162804377, "tokens_seen": 430309376 }, { "epoch": 0.13, "learning_rate": 0.00043918756268806423, "loss": 2.8373, "theoretical_loss": 3.979433192711592, "tokens_seen": 430374912 }, { "epoch": 0.13, "learning_rate": 0.0004391775325977934, "loss": 3.0308, "theoretical_loss": 3.9793662356709407, "tokens_seen": 430440448 }, { "epoch": 0.13, "learning_rate": 0.0004391675025075226, "loss": 2.8456, "theoretical_loss": 3.979299291677894, "tokens_seen": 430505984 }, { "epoch": 0.13, "learning_rate": 0.00043915747241725177, "loss": 2.863, "theoretical_loss": 3.9792323607279223, "tokens_seen": 430571520 }, { "epoch": 0.13, "learning_rate": 0.00043914744232698095, "loss": 2.9346, "theoretical_loss": 3.979165442816501, "tokens_seen": 430637056 }, { "epoch": 0.13, "learning_rate": 0.0004391374122367102, "loss": 3.1467, "theoretical_loss": 3.9790985379391053, "tokens_seen": 430702592 }, { "epoch": 0.13, "learning_rate": 0.0004391273821464393, "loss": 2.9044, "theoretical_loss": 3.9790316460912143, "tokens_seen": 430768128 }, { "epoch": 0.13, "learning_rate": 0.00043911735205616855, "loss": 2.9496, "theoretical_loss": 3.978964767268309, "tokens_seen": 430833664 }, { "epoch": 0.13, "objective/train/docs_used": 256338, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.68479585647583, "objective/train/theoretical_loss": 3.9788979014658716, "objective/train/tokens_used": 451359200, "theoretical_loss": 3.9788979014658716, "tokens_seen": 430899200 }, { "epoch": 0.13, "learning_rate": 0.0004391073219658977, "loss": 2.7706, "theoretical_loss": 3.9788979014658716, "tokens_seen": 430899200 }, { "epoch": 0.13, "learning_rate": 0.0004390972918756269, "loss": 2.7733, "theoretical_loss": 3.978831048679389, "tokens_seen": 430964736 }, { "epoch": 0.13, "learning_rate": 0.0004390872617853561, "loss": 3.0516, "theoretical_loss": 3.978764208904347, "tokens_seen": 431030272 }, { "epoch": 0.13, "learning_rate": 0.0004390772316950853, "loss": 2.9285, "theoretical_loss": 3.9786973821362377, "tokens_seen": 431095808 }, { "epoch": 0.13, "learning_rate": 0.00043906720160481446, "loss": 2.9211, "theoretical_loss": 3.9786305683705505, "tokens_seen": 431161344 }, { "epoch": 0.13, "learning_rate": 0.00043905717151454364, "loss": 2.9285, "theoretical_loss": 3.978563767602782, "tokens_seen": 431226880 }, { "epoch": 0.13, "learning_rate": 0.0004390471414242728, "loss": 2.6945, "theoretical_loss": 3.978496979828428, "tokens_seen": 431292416 }, { "epoch": 0.13, "learning_rate": 0.00043903711133400205, "loss": 2.889, "theoretical_loss": 3.9784302050429874, "tokens_seen": 431357952 }, { "epoch": 0.13, "learning_rate": 0.0004390270812437312, "loss": 3.1345, "theoretical_loss": 3.978363443241961, "tokens_seen": 431423488 }, { "epoch": 0.13, "learning_rate": 0.0004390170511534604, "loss": 2.9283, "theoretical_loss": 3.9782966944208518, "tokens_seen": 431489024 }, { "epoch": 0.13, "learning_rate": 0.0004390070210631896, "loss": 3.0917, "theoretical_loss": 3.9782299585751666, "tokens_seen": 431554560 }, { "epoch": 0.13, "learning_rate": 0.0004389969909729188, "loss": 2.7283, "theoretical_loss": 3.9781632357004124, "tokens_seen": 431620096 }, { "epoch": 0.13, "learning_rate": 0.00043898696088264796, "loss": 2.92, "theoretical_loss": 3.9780965257920995, "tokens_seen": 431685632 }, { "epoch": 0.13, "learning_rate": 0.00043897693079237714, "loss": 3.0776, "theoretical_loss": 3.9780298288457394, "tokens_seen": 431751168 }, { "epoch": 0.13, "learning_rate": 0.0004389669007021063, "loss": 2.8729, "theoretical_loss": 3.9779631448568473, "tokens_seen": 431816704 }, { "epoch": 0.13, "learning_rate": 0.00043895687061183556, "loss": 2.9113, "theoretical_loss": 3.977896473820941, "tokens_seen": 431882240 }, { "epoch": 0.13, "learning_rate": 0.0004389468405215647, "loss": 2.8545, "theoretical_loss": 3.9778298157335374, "tokens_seen": 431947776 }, { "epoch": 0.13, "learning_rate": 0.0004389368104312939, "loss": 3.0507, "theoretical_loss": 3.977763170590159, "tokens_seen": 432013312 }, { "epoch": 0.13, "learning_rate": 0.00043892678034102305, "loss": 2.8307, "theoretical_loss": 3.9776965383863288, "tokens_seen": 432078848 }, { "epoch": 0.13, "learning_rate": 0.0004389167502507523, "loss": 2.8236, "theoretical_loss": 3.9776299191175726, "tokens_seen": 432144384 }, { "epoch": 0.13, "learning_rate": 0.00043890672016048146, "loss": 2.9224, "theoretical_loss": 3.977563312779418, "tokens_seen": 432209920 }, { "epoch": 0.13, "learning_rate": 0.00043889669007021064, "loss": 2.8145, "theoretical_loss": 3.977496719367396, "tokens_seen": 432275456 }, { "epoch": 0.13, "learning_rate": 0.0004388866599799398, "loss": 2.8554, "theoretical_loss": 3.9774301388770374, "tokens_seen": 432340992 }, { "epoch": 0.13, "learning_rate": 0.000438876629889669, "loss": 3.0139, "theoretical_loss": 3.9773635713038784, "tokens_seen": 432406528 }, { "epoch": 0.13, "learning_rate": 0.0004388665997993982, "loss": 2.9057, "theoretical_loss": 3.9772970166434547, "tokens_seen": 432472064 }, { "epoch": 0.13, "objective/train/docs_used": 257001, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.186760187149048, "objective/train/theoretical_loss": 3.9772304748913054, "objective/train/tokens_used": 452997600, "theoretical_loss": 3.9772304748913054, "tokens_seen": 432537600 }, { "epoch": 0.13, "learning_rate": 0.0004388565697091274, "loss": 2.946, "theoretical_loss": 3.9772304748913054, "tokens_seen": 432537600 }, { "epoch": 0.13, "learning_rate": 0.00043884653961885655, "loss": 2.8407, "theoretical_loss": 3.977163946042973, "tokens_seen": 432603136 }, { "epoch": 0.13, "learning_rate": 0.0004388365095285858, "loss": 2.9356, "theoretical_loss": 3.977097430093999, "tokens_seen": 432668672 }, { "epoch": 0.13, "learning_rate": 0.00043882647943831497, "loss": 2.9862, "theoretical_loss": 3.97703092703993, "tokens_seen": 432734208 }, { "epoch": 0.13, "learning_rate": 0.00043881644934804415, "loss": 2.9973, "theoretical_loss": 3.9769644368763135, "tokens_seen": 432799744 }, { "epoch": 0.13, "learning_rate": 0.00043880641925777333, "loss": 3.1139, "theoretical_loss": 3.9768979595987, "tokens_seen": 432865280 }, { "epoch": 0.13, "learning_rate": 0.0004387963891675025, "loss": 3.0764, "theoretical_loss": 3.9768314952026422, "tokens_seen": 432930816 }, { "epoch": 0.13, "learning_rate": 0.0004387863590772317, "loss": 2.8346, "theoretical_loss": 3.9767650436836934, "tokens_seen": 432996352 }, { "epoch": 0.13, "learning_rate": 0.0004387763289869609, "loss": 2.9051, "theoretical_loss": 3.976698605037411, "tokens_seen": 433061888 }, { "epoch": 0.13, "learning_rate": 0.00043876629889669005, "loss": 2.9545, "theoretical_loss": 3.976632179259354, "tokens_seen": 433127424 }, { "epoch": 0.13, "learning_rate": 0.0004387562688064193, "loss": 3.116, "theoretical_loss": 3.976565766345083, "tokens_seen": 433192960 }, { "epoch": 0.13, "learning_rate": 0.0004387462387161484, "loss": 2.8026, "theoretical_loss": 3.9764993662901613, "tokens_seen": 433258496 }, { "epoch": 0.13, "learning_rate": 0.00043873620862587765, "loss": 2.9205, "theoretical_loss": 3.976432979090155, "tokens_seen": 433324032 }, { "epoch": 0.13, "learning_rate": 0.00043872617853560683, "loss": 2.9889, "theoretical_loss": 3.9763666047406314, "tokens_seen": 433389568 }, { "epoch": 0.13, "learning_rate": 0.000438716148445336, "loss": 2.8799, "theoretical_loss": 3.9763002432371604, "tokens_seen": 433455104 }, { "epoch": 0.13, "learning_rate": 0.0004387061183550652, "loss": 3.0946, "theoretical_loss": 3.9762338945753144, "tokens_seen": 433520640 }, { "epoch": 0.13, "learning_rate": 0.00043869608826479443, "loss": 2.9393, "theoretical_loss": 3.9761675587506673, "tokens_seen": 433586176 }, { "epoch": 0.13, "learning_rate": 0.00043868605817452356, "loss": 3.1171, "theoretical_loss": 3.9761012357587955, "tokens_seen": 433651712 }, { "epoch": 0.13, "learning_rate": 0.0004386760280842528, "loss": 2.8962, "theoretical_loss": 3.976034925595278, "tokens_seen": 433717248 }, { "epoch": 0.13, "learning_rate": 0.0004386659979939819, "loss": 2.908, "theoretical_loss": 3.975968628255696, "tokens_seen": 433782784 }, { "epoch": 0.13, "learning_rate": 0.00043865596790371115, "loss": 3.0901, "theoretical_loss": 3.9759023437356316, "tokens_seen": 433848320 }, { "epoch": 0.13, "learning_rate": 0.00043864593781344033, "loss": 2.8724, "theoretical_loss": 3.9758360720306705, "tokens_seen": 433913856 }, { "epoch": 0.13, "learning_rate": 0.0004386359077231695, "loss": 3.1435, "theoretical_loss": 3.9757698131364005, "tokens_seen": 433979392 }, { "epoch": 0.13, "learning_rate": 0.0004386258776328987, "loss": 2.8047, "theoretical_loss": 3.9757035670484107, "tokens_seen": 434044928 }, { "epoch": 0.13, "learning_rate": 0.0004386158475426279, "loss": 2.782, "theoretical_loss": 3.9756373337622932, "tokens_seen": 434110464 }, { "epoch": 0.13, "objective/train/docs_used": 257982, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.2712156772613525, "objective/train/theoretical_loss": 3.975571113273642, "objective/train/tokens_used": 454636000, "theoretical_loss": 3.975571113273642, "tokens_seen": 434176000 }, { "epoch": 0.13, "learning_rate": 0.00043860581745235706, "loss": 2.8521, "theoretical_loss": 3.975571113273642, "tokens_seen": 434176000 }, { "epoch": 0.13, "learning_rate": 0.0004385957873620863, "loss": 2.785, "theoretical_loss": 3.975504905578053, "tokens_seen": 434241536 }, { "epoch": 0.13, "learning_rate": 0.0004385857572718154, "loss": 2.8655, "theoretical_loss": 3.975438710671125, "tokens_seen": 434307072 }, { "epoch": 0.13, "learning_rate": 0.00043857572718154466, "loss": 2.9464, "theoretical_loss": 3.9753725285484576, "tokens_seen": 434372608 }, { "epoch": 0.13, "learning_rate": 0.0004385656970912738, "loss": 3.0666, "theoretical_loss": 3.9753063592056552, "tokens_seen": 434438144 }, { "epoch": 0.13, "learning_rate": 0.000438555667001003, "loss": 2.9942, "theoretical_loss": 3.9752402026383207, "tokens_seen": 434503680 }, { "epoch": 0.13, "learning_rate": 0.0004385456369107322, "loss": 3.035, "theoretical_loss": 3.9751740588420623, "tokens_seen": 434569216 }, { "epoch": 0.13, "learning_rate": 0.0004385356068204614, "loss": 2.9534, "theoretical_loss": 3.9751079278124886, "tokens_seen": 434634752 }, { "epoch": 0.13, "learning_rate": 0.00043852557673019056, "loss": 3.0763, "theoretical_loss": 3.9750418095452122, "tokens_seen": 434700288 }, { "epoch": 0.13, "learning_rate": 0.0004385155466399198, "loss": 2.9295, "theoretical_loss": 3.9749757040358453, "tokens_seen": 434765824 }, { "epoch": 0.13, "learning_rate": 0.0004385055165496489, "loss": 2.9565, "theoretical_loss": 3.9749096112800046, "tokens_seen": 434831360 }, { "epoch": 0.13, "learning_rate": 0.00043849548645937816, "loss": 2.7425, "theoretical_loss": 3.974843531273307, "tokens_seen": 434896896 }, { "epoch": 0.13, "learning_rate": 0.0004384854563691073, "loss": 2.8579, "theoretical_loss": 3.9747774640113738, "tokens_seen": 434962432 }, { "epoch": 0.13, "learning_rate": 0.0004384754262788365, "loss": 2.7193, "theoretical_loss": 3.974711409489826, "tokens_seen": 435027968 }, { "epoch": 0.13, "learning_rate": 0.0004384653961885657, "loss": 2.8602, "theoretical_loss": 3.974645367704289, "tokens_seen": 435093504 }, { "epoch": 0.13, "learning_rate": 0.0004384553660982949, "loss": 2.9948, "theoretical_loss": 3.9745793386503885, "tokens_seen": 435159040 }, { "epoch": 0.13, "learning_rate": 0.00043844533600802407, "loss": 2.9632, "theoretical_loss": 3.974513322323754, "tokens_seen": 435224576 }, { "epoch": 0.13, "learning_rate": 0.00043843530591775325, "loss": 2.7926, "theoretical_loss": 3.9744473187200158, "tokens_seen": 435290112 }, { "epoch": 0.13, "learning_rate": 0.0004384252758274825, "loss": 3.1408, "theoretical_loss": 3.974381327834807, "tokens_seen": 435355648 }, { "epoch": 0.13, "learning_rate": 0.00043841524573721166, "loss": 3.0636, "theoretical_loss": 3.974315349663763, "tokens_seen": 435421184 }, { "epoch": 0.13, "learning_rate": 0.00043840521564694084, "loss": 2.8573, "theoretical_loss": 3.9742493842025213, "tokens_seen": 435486720 }, { "epoch": 0.13, "learning_rate": 0.00043839518555667, "loss": 2.7695, "theoretical_loss": 3.974183431446721, "tokens_seen": 435552256 }, { "epoch": 0.13, "learning_rate": 0.0004383851554663992, "loss": 2.8558, "theoretical_loss": 3.974117491392004, "tokens_seen": 435617792 }, { "epoch": 0.13, "learning_rate": 0.0004383751253761284, "loss": 2.7744, "theoretical_loss": 3.9740515640340135, "tokens_seen": 435683328 }, { "epoch": 0.13, "learning_rate": 0.0004383650952858576, "loss": 3.0929, "theoretical_loss": 3.9739856493683967, "tokens_seen": 435748864 }, { "epoch": 0.13, "objective/train/docs_used": 258654, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9247915744781494, "objective/train/theoretical_loss": 3.973919747390801, "objective/train/tokens_used": 456274400, "theoretical_loss": 3.973919747390801, "tokens_seen": 435814400 }, { "epoch": 0.13, "learning_rate": 0.00043835506519558675, "loss": 3.015, "theoretical_loss": 3.973919747390801, "tokens_seen": 435814400 }, { "epoch": 0.13, "learning_rate": 0.000438345035105316, "loss": 3.172, "theoretical_loss": 3.973853858096876, "tokens_seen": 435879936 }, { "epoch": 0.13, "learning_rate": 0.00043833500501504517, "loss": 2.922, "theoretical_loss": 3.9737879814822747, "tokens_seen": 435945472 }, { "epoch": 0.13, "learning_rate": 0.00043832497492477435, "loss": 2.889, "theoretical_loss": 3.973722117542652, "tokens_seen": 436011008 }, { "epoch": 0.13, "learning_rate": 0.00043831494483450353, "loss": 2.9686, "theoretical_loss": 3.9736562662736636, "tokens_seen": 436076544 }, { "epoch": 0.13, "learning_rate": 0.0004383049147442327, "loss": 2.9584, "theoretical_loss": 3.9735904276709695, "tokens_seen": 436142080 }, { "epoch": 0.13, "learning_rate": 0.0004382948846539619, "loss": 2.9236, "theoretical_loss": 3.9735246017302295, "tokens_seen": 436207616 }, { "epoch": 0.13, "learning_rate": 0.0004382848545636911, "loss": 3.0206, "theoretical_loss": 3.9734587884471075, "tokens_seen": 436273152 }, { "epoch": 0.13, "learning_rate": 0.00043827482447342025, "loss": 2.9689, "theoretical_loss": 3.9733929878172685, "tokens_seen": 436338688 }, { "epoch": 0.13, "learning_rate": 0.0004382647943831495, "loss": 2.8583, "theoretical_loss": 3.9733271998363797, "tokens_seen": 436404224 }, { "epoch": 0.13, "learning_rate": 0.0004382547642928786, "loss": 3.0176, "theoretical_loss": 3.9732614245001106, "tokens_seen": 436469760 }, { "epoch": 0.13, "learning_rate": 0.00043824473420260785, "loss": 2.8963, "theoretical_loss": 3.9731956618041333, "tokens_seen": 436535296 }, { "epoch": 0.13, "learning_rate": 0.00043823470411233703, "loss": 2.9904, "theoretical_loss": 3.9731299117441203, "tokens_seen": 436600832 }, { "epoch": 0.13, "learning_rate": 0.0004382246740220662, "loss": 3.0376, "theoretical_loss": 3.9730641743157493, "tokens_seen": 436666368 }, { "epoch": 0.13, "learning_rate": 0.0004382146439317954, "loss": 2.8604, "theoretical_loss": 3.972998449514697, "tokens_seen": 436731904 }, { "epoch": 0.13, "learning_rate": 0.00043820461384152463, "loss": 2.7958, "theoretical_loss": 3.9729327373366443, "tokens_seen": 436797440 }, { "epoch": 0.13, "learning_rate": 0.00043819458375125376, "loss": 2.827, "theoretical_loss": 3.972867037777273, "tokens_seen": 436862976 }, { "epoch": 0.13, "learning_rate": 0.000438184553660983, "loss": 2.9757, "theoretical_loss": 3.9728013508322673, "tokens_seen": 436928512 }, { "epoch": 0.13, "learning_rate": 0.0004381745235707121, "loss": 2.8648, "theoretical_loss": 3.9727356764973143, "tokens_seen": 436994048 }, { "epoch": 0.13, "learning_rate": 0.00043816449348044135, "loss": 2.9147, "theoretical_loss": 3.972670014768102, "tokens_seen": 437059584 }, { "epoch": 0.13, "learning_rate": 0.00043815446339017053, "loss": 2.9013, "theoretical_loss": 3.972604365640322, "tokens_seen": 437125120 }, { "epoch": 0.13, "learning_rate": 0.0004381444332998997, "loss": 2.9928, "theoretical_loss": 3.972538729109666, "tokens_seen": 437190656 }, { "epoch": 0.13, "learning_rate": 0.0004381344032096289, "loss": 2.9384, "theoretical_loss": 3.97247310517183, "tokens_seen": 437256192 }, { "epoch": 0.13, "learning_rate": 0.0004381243731193581, "loss": 3.2878, "theoretical_loss": 3.972407493822511, "tokens_seen": 437321728 }, { "epoch": 0.13, "learning_rate": 0.00043811434302908726, "loss": 2.6546, "theoretical_loss": 3.972341895057408, "tokens_seen": 437387264 }, { "epoch": 0.13, "objective/train/docs_used": 259916, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.601888418197632, "objective/train/theoretical_loss": 3.972276308872222, "objective/train/tokens_used": 457912800, "theoretical_loss": 3.972276308872222, "tokens_seen": 437452800 }, { "epoch": 0.13, "learning_rate": 0.0004381043129388165, "loss": 2.7597, "theoretical_loss": 3.972276308872222, "tokens_seen": 437452800 }, { "epoch": 0.13, "learning_rate": 0.0004380942828485456, "loss": 2.8085, "theoretical_loss": 3.972210735262657, "tokens_seen": 437518336 }, { "epoch": 0.13, "learning_rate": 0.00043808425275827486, "loss": 3.1744, "theoretical_loss": 3.9721451742244183, "tokens_seen": 437583872 }, { "epoch": 0.13, "learning_rate": 0.000438074222668004, "loss": 3.0446, "theoretical_loss": 3.9720796257532136, "tokens_seen": 437649408 }, { "epoch": 0.13, "learning_rate": 0.0004380641925777332, "loss": 2.8964, "theoretical_loss": 3.9720140898447527, "tokens_seen": 437714944 }, { "epoch": 0.13, "learning_rate": 0.0004380541624874624, "loss": 3.0913, "theoretical_loss": 3.9719485664947483, "tokens_seen": 437780480 }, { "epoch": 0.13, "learning_rate": 0.0004380441323971916, "loss": 2.9713, "theoretical_loss": 3.971883055698913, "tokens_seen": 437846016 }, { "epoch": 0.13, "learning_rate": 0.00043803410230692076, "loss": 3.012, "theoretical_loss": 3.9718175574529635, "tokens_seen": 437911552 }, { "epoch": 0.13, "learning_rate": 0.00043802407221665, "loss": 2.8964, "theoretical_loss": 3.9717520717526176, "tokens_seen": 437977088 }, { "epoch": 0.13, "learning_rate": 0.0004380140421263791, "loss": 3.1043, "theoretical_loss": 3.971686598593596, "tokens_seen": 438042624 }, { "epoch": 0.13, "learning_rate": 0.00043800401203610836, "loss": 2.9459, "theoretical_loss": 3.971621137971622, "tokens_seen": 438108160 }, { "epoch": 0.13, "learning_rate": 0.0004379939819458375, "loss": 3.0117, "theoretical_loss": 3.9715556898824182, "tokens_seen": 438173696 }, { "epoch": 0.13, "learning_rate": 0.0004379839518555667, "loss": 3.0628, "theoretical_loss": 3.971490254321713, "tokens_seen": 438239232 }, { "epoch": 0.13, "learning_rate": 0.0004379739217652959, "loss": 3.005, "theoretical_loss": 3.9714248312852343, "tokens_seen": 438304768 }, { "epoch": 0.13, "learning_rate": 0.0004379638916750251, "loss": 3.2229, "theoretical_loss": 3.971359420768712, "tokens_seen": 438370304 }, { "epoch": 0.13, "learning_rate": 0.00043795386158475427, "loss": 3.0165, "theoretical_loss": 3.9712940227678803, "tokens_seen": 438435840 }, { "epoch": 0.13, "learning_rate": 0.00043794383149448345, "loss": 2.9107, "theoretical_loss": 3.9712286372784735, "tokens_seen": 438501376 }, { "epoch": 0.13, "learning_rate": 0.00043793380140421263, "loss": 3.1229, "theoretical_loss": 3.971163264296229, "tokens_seen": 438566912 }, { "epoch": 0.13, "learning_rate": 0.00043792377131394186, "loss": 2.9839, "theoretical_loss": 3.971097903816886, "tokens_seen": 438632448 }, { "epoch": 0.13, "learning_rate": 0.000437913741223671, "loss": 2.8638, "theoretical_loss": 3.9710325558361856, "tokens_seen": 438697984 }, { "epoch": 0.13, "learning_rate": 0.0004379037111334002, "loss": 2.8033, "theoretical_loss": 3.9709672203498707, "tokens_seen": 438763520 }, { "epoch": 0.13, "learning_rate": 0.00043789368104312935, "loss": 2.9359, "theoretical_loss": 3.970901897353688, "tokens_seen": 438829056 }, { "epoch": 0.13, "learning_rate": 0.0004378836509528586, "loss": 2.8778, "theoretical_loss": 3.970836586843383, "tokens_seen": 438894592 }, { "epoch": 0.13, "learning_rate": 0.00043787362086258777, "loss": 3.0624, "theoretical_loss": 3.9707712888147073, "tokens_seen": 438960128 }, { "epoch": 0.13, "learning_rate": 0.00043786359077231695, "loss": 2.7369, "theoretical_loss": 3.9707060032634107, "tokens_seen": 439025664 }, { "epoch": 0.13, "objective/train/docs_used": 260524, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9223694801330566, "objective/train/theoretical_loss": 3.9706407301852487, "objective/train/tokens_used": 459551200, "theoretical_loss": 3.9706407301852487, "tokens_seen": 439091200 }, { "epoch": 0.13, "learning_rate": 0.00043785356068204613, "loss": 2.7099, "theoretical_loss": 3.9706407301852487, "tokens_seen": 439091200 }, { "epoch": 0.13, "learning_rate": 0.00043784353059177537, "loss": 2.9097, "theoretical_loss": 3.9705754695759756, "tokens_seen": 439156736 }, { "epoch": 0.13, "learning_rate": 0.0004378335005015045, "loss": 2.7865, "theoretical_loss": 3.9705102214313506, "tokens_seen": 439222272 }, { "epoch": 0.13, "learning_rate": 0.00043782347041123373, "loss": 2.9523, "theoretical_loss": 3.9704449857471324, "tokens_seen": 439287808 }, { "epoch": 0.13, "learning_rate": 0.00043781344032096286, "loss": 3.0496, "theoretical_loss": 3.970379762519084, "tokens_seen": 439353344 }, { "epoch": 0.13, "learning_rate": 0.0004378034102306921, "loss": 2.9437, "theoretical_loss": 3.9703145517429697, "tokens_seen": 439418880 }, { "epoch": 0.13, "learning_rate": 0.00043779338014042127, "loss": 3.0335, "theoretical_loss": 3.9702493534145544, "tokens_seen": 439484416 }, { "epoch": 0.13, "learning_rate": 0.00043778335005015045, "loss": 3.0759, "theoretical_loss": 3.9701841675296077, "tokens_seen": 439549952 }, { "epoch": 0.13, "learning_rate": 0.00043777331995987963, "loss": 3.2439, "theoretical_loss": 3.970118994083899, "tokens_seen": 439615488 }, { "epoch": 0.13, "learning_rate": 0.0004377632898696088, "loss": 3.0867, "theoretical_loss": 3.9700538330732016, "tokens_seen": 439681024 }, { "epoch": 0.13, "learning_rate": 0.000437753259779338, "loss": 3.1688, "theoretical_loss": 3.969988684493289, "tokens_seen": 439746560 }, { "epoch": 0.13, "learning_rate": 0.00043774322968906723, "loss": 3.0956, "theoretical_loss": 3.9699235483399384, "tokens_seen": 439812096 }, { "epoch": 0.13, "learning_rate": 0.00043773319959879636, "loss": 2.8337, "theoretical_loss": 3.969858424608928, "tokens_seen": 439877632 }, { "epoch": 0.13, "learning_rate": 0.0004377231695085256, "loss": 3.0612, "theoretical_loss": 3.969793313296039, "tokens_seen": 439943168 }, { "epoch": 0.13, "learning_rate": 0.0004377131394182547, "loss": 3.1136, "theoretical_loss": 3.969728214397053, "tokens_seen": 440008704 }, { "epoch": 0.13, "learning_rate": 0.00043770310932798396, "loss": 2.9982, "theoretical_loss": 3.969663127907756, "tokens_seen": 440074240 }, { "epoch": 0.13, "learning_rate": 0.00043769307923771314, "loss": 3.004, "theoretical_loss": 3.969598053823934, "tokens_seen": 440139776 }, { "epoch": 0.13, "learning_rate": 0.0004376830491474423, "loss": 2.9279, "theoretical_loss": 3.969532992141377, "tokens_seen": 440205312 }, { "epoch": 0.13, "learning_rate": 0.00043767301905717155, "loss": 2.9182, "theoretical_loss": 3.969467942855875, "tokens_seen": 440270848 }, { "epoch": 0.13, "learning_rate": 0.00043766298896690073, "loss": 3.1425, "theoretical_loss": 3.969402905963221, "tokens_seen": 440336384 }, { "epoch": 0.13, "learning_rate": 0.0004376529588766299, "loss": 3.2916, "theoretical_loss": 3.9693378814592104, "tokens_seen": 440401920 }, { "epoch": 0.13, "learning_rate": 0.0004376429287863591, "loss": 3.0597, "theoretical_loss": 3.96927286933964, "tokens_seen": 440467456 }, { "epoch": 0.13, "learning_rate": 0.0004376328986960883, "loss": 2.8762, "theoretical_loss": 3.969207869600309, "tokens_seen": 440532992 }, { "epoch": 0.13, "learning_rate": 0.00043762286860581746, "loss": 3.1591, "theoretical_loss": 3.969142882237019, "tokens_seen": 440598528 }, { "epoch": 0.13, "learning_rate": 0.0004376128385155467, "loss": 3.0577, "theoretical_loss": 3.9690779072455733, "tokens_seen": 440664064 }, { "epoch": 0.13, "objective/train/docs_used": 261739, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.807461977005005, "objective/train/theoretical_loss": 3.969012944621776, "objective/train/tokens_used": 461189600, "theoretical_loss": 3.969012944621776, "tokens_seen": 440729600 }, { "epoch": 0.13, "learning_rate": 0.0004376028084252758, "loss": 2.6719, "theoretical_loss": 3.969012944621776, "tokens_seen": 440729600 }, { "epoch": 0.13, "learning_rate": 0.00043759277833500506, "loss": 2.9454, "theoretical_loss": 3.968947994361436, "tokens_seen": 440795136 }, { "epoch": 0.13, "learning_rate": 0.0004375827482447342, "loss": 2.6968, "theoretical_loss": 3.9688830564603625, "tokens_seen": 440860672 }, { "epoch": 0.13, "learning_rate": 0.0004375727181544634, "loss": 2.9212, "theoretical_loss": 3.9688181309143657, "tokens_seen": 440926208 }, { "epoch": 0.13, "learning_rate": 0.0004375626880641926, "loss": 2.8568, "theoretical_loss": 3.9687532177192604, "tokens_seen": 440991744 }, { "epoch": 0.13, "learning_rate": 0.0004375526579739218, "loss": 3.0616, "theoretical_loss": 3.9686883168708613, "tokens_seen": 441057280 }, { "epoch": 0.13, "learning_rate": 0.00043754262788365096, "loss": 2.9143, "theoretical_loss": 3.968623428364986, "tokens_seen": 441122816 }, { "epoch": 0.13, "learning_rate": 0.0004375325977933802, "loss": 2.9528, "theoretical_loss": 3.9685585521974556, "tokens_seen": 441188352 }, { "epoch": 0.13, "learning_rate": 0.0004375225677031093, "loss": 2.9707, "theoretical_loss": 3.968493688364089, "tokens_seen": 441253888 }, { "epoch": 0.13, "learning_rate": 0.00043751253761283856, "loss": 3.0977, "theoretical_loss": 3.968428836860712, "tokens_seen": 441319424 }, { "epoch": 0.13, "learning_rate": 0.0004375025075225677, "loss": 2.7931, "theoretical_loss": 3.96836399768315, "tokens_seen": 441384960 }, { "epoch": 0.13, "learning_rate": 0.0004374924774322969, "loss": 3.0323, "theoretical_loss": 3.96829917082723, "tokens_seen": 441450496 }, { "epoch": 0.13, "learning_rate": 0.0004374824473420261, "loss": 3.1413, "theoretical_loss": 3.9682343562887823, "tokens_seen": 441516032 }, { "epoch": 0.13, "learning_rate": 0.0004374724172517553, "loss": 3.2888, "theoretical_loss": 3.968169554063639, "tokens_seen": 441581568 }, { "epoch": 0.13, "learning_rate": 0.00043746238716148447, "loss": 3.019, "theoretical_loss": 3.9681047641476326, "tokens_seen": 441647104 }, { "epoch": 0.13, "learning_rate": 0.00043745235707121365, "loss": 2.8898, "theoretical_loss": 3.968039986536601, "tokens_seen": 441712640 }, { "epoch": 0.13, "learning_rate": 0.00043744232698094283, "loss": 3.1168, "theoretical_loss": 3.9679752212263795, "tokens_seen": 441778176 }, { "epoch": 0.13, "learning_rate": 0.00043743229689067206, "loss": 2.8963, "theoretical_loss": 3.9679104682128106, "tokens_seen": 441843712 }, { "epoch": 0.13, "learning_rate": 0.0004374222668004012, "loss": 2.9959, "theoretical_loss": 3.967845727491735, "tokens_seen": 441909248 }, { "epoch": 0.13, "learning_rate": 0.0004374122367101304, "loss": 2.9724, "theoretical_loss": 3.9677809990589967, "tokens_seen": 441974784 }, { "epoch": 0.13, "learning_rate": 0.00043740220661985955, "loss": 2.7693, "theoretical_loss": 3.967716282910441, "tokens_seen": 442040320 }, { "epoch": 0.13, "learning_rate": 0.0004373921765295888, "loss": 2.8861, "theoretical_loss": 3.9676515790419176, "tokens_seen": 442105856 }, { "epoch": 0.13, "learning_rate": 0.00043738214643931797, "loss": 2.9332, "theoretical_loss": 3.967586887449275, "tokens_seen": 442171392 }, { "epoch": 0.13, "learning_rate": 0.00043737211634904715, "loss": 2.9028, "theoretical_loss": 3.967522208128366, "tokens_seen": 442236928 }, { "epoch": 0.13, "learning_rate": 0.00043736208625877633, "loss": 3.0509, "theoretical_loss": 3.967457541075044, "tokens_seen": 442302464 }, { "epoch": 0.13, "objective/train/docs_used": 262441, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8256170749664307, "objective/train/theoretical_loss": 3.9673928862851655, "objective/train/tokens_used": 462828000, "theoretical_loss": 3.9673928862851655, "tokens_seen": 442368000 }, { "epoch": 0.13, "learning_rate": 0.00043735205616850557, "loss": 2.7654, "theoretical_loss": 3.9673928862851655, "tokens_seen": 442368000 }, { "epoch": 0.13, "learning_rate": 0.0004373420260782347, "loss": 2.9835, "theoretical_loss": 3.9673282437545883, "tokens_seen": 442433536 }, { "epoch": 0.13, "learning_rate": 0.00043733199598796393, "loss": 3.1724, "theoretical_loss": 3.9672636134791737, "tokens_seen": 442499072 }, { "epoch": 0.13, "learning_rate": 0.00043732196589769306, "loss": 2.9039, "theoretical_loss": 3.967198995454782, "tokens_seen": 442564608 }, { "epoch": 0.13, "learning_rate": 0.0004373119358074223, "loss": 2.7732, "theoretical_loss": 3.9671343896772786, "tokens_seen": 442630144 }, { "epoch": 0.13, "learning_rate": 0.00043730190571715147, "loss": 2.7082, "theoretical_loss": 3.9670697961425287, "tokens_seen": 442695680 }, { "epoch": 0.13, "learning_rate": 0.00043729187562688065, "loss": 3.2087, "theoretical_loss": 3.967005214846401, "tokens_seen": 442761216 }, { "epoch": 0.13, "learning_rate": 0.00043728184553660983, "loss": 3.0608, "theoretical_loss": 3.9669406457847654, "tokens_seen": 442826752 }, { "epoch": 0.13, "learning_rate": 0.000437271815446339, "loss": 2.9636, "theoretical_loss": 3.9668760889534944, "tokens_seen": 442892288 }, { "epoch": 0.13, "learning_rate": 0.0004372617853560682, "loss": 3.1686, "theoretical_loss": 3.9668115443484617, "tokens_seen": 442957824 }, { "epoch": 0.13, "learning_rate": 0.00043725175526579743, "loss": 2.8636, "theoretical_loss": 3.966747011965544, "tokens_seen": 443023360 }, { "epoch": 0.13, "learning_rate": 0.00043724172517552656, "loss": 2.7879, "theoretical_loss": 3.9666824918006185, "tokens_seen": 443088896 }, { "epoch": 0.13, "learning_rate": 0.0004372316950852558, "loss": 2.8745, "theoretical_loss": 3.9666179838495657, "tokens_seen": 443154432 }, { "epoch": 0.13, "learning_rate": 0.0004372216649949849, "loss": 2.9572, "theoretical_loss": 3.9665534881082687, "tokens_seen": 443219968 }, { "epoch": 0.13, "learning_rate": 0.00043721163490471416, "loss": 2.9991, "theoretical_loss": 3.96648900457261, "tokens_seen": 443285504 }, { "epoch": 0.13, "learning_rate": 0.00043720160481444334, "loss": 3.0013, "theoretical_loss": 3.966424533238477, "tokens_seen": 443351040 }, { "epoch": 0.13, "learning_rate": 0.0004371915747241725, "loss": 2.9459, "theoretical_loss": 3.9663600741017575, "tokens_seen": 443416576 }, { "epoch": 0.13, "learning_rate": 0.0004371815446339017, "loss": 2.9219, "theoretical_loss": 3.966295627158341, "tokens_seen": 443482112 }, { "epoch": 0.13, "learning_rate": 0.00043717151454363093, "loss": 2.8067, "theoretical_loss": 3.9662311924041203, "tokens_seen": 443547648 }, { "epoch": 0.13, "learning_rate": 0.00043716148445336006, "loss": 3.0382, "theoretical_loss": 3.96616676983499, "tokens_seen": 443613184 }, { "epoch": 0.13, "learning_rate": 0.0004371514543630893, "loss": 3.0499, "theoretical_loss": 3.966102359446845, "tokens_seen": 443678720 }, { "epoch": 0.13, "learning_rate": 0.0004371414242728184, "loss": 2.7958, "theoretical_loss": 3.9660379612355836, "tokens_seen": 443744256 }, { "epoch": 0.13, "learning_rate": 0.00043713139418254766, "loss": 2.9751, "theoretical_loss": 3.9659735751971064, "tokens_seen": 443809792 }, { "epoch": 0.13, "learning_rate": 0.00043712136409227684, "loss": 2.864, "theoretical_loss": 3.965909201327315, "tokens_seen": 443875328 }, { "epoch": 0.13, "learning_rate": 0.000437111334002006, "loss": 2.9858, "theoretical_loss": 3.9658448396221138, "tokens_seen": 443940864 }, { "epoch": 0.13, "objective/train/docs_used": 263839, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.407069444656372, "objective/train/theoretical_loss": 3.9657804900774085, "objective/train/tokens_used": 464466400, "theoretical_loss": 3.9657804900774085, "tokens_seen": 444006400 }, { "epoch": 0.13, "learning_rate": 0.0004371013039117352, "loss": 2.6557, "theoretical_loss": 3.9657804900774085, "tokens_seen": 444006400 }, { "epoch": 0.13, "learning_rate": 0.0004370912738214644, "loss": 2.9231, "theoretical_loss": 3.9657161526891076, "tokens_seen": 444071936 }, { "epoch": 0.13, "learning_rate": 0.00043708124373119356, "loss": 3.0561, "theoretical_loss": 3.9656518274531205, "tokens_seen": 444137472 }, { "epoch": 0.13, "learning_rate": 0.0004370712136409228, "loss": 2.9919, "theoretical_loss": 3.9655875143653594, "tokens_seen": 444203008 }, { "epoch": 0.13, "learning_rate": 0.0004370611835506519, "loss": 3.0073, "theoretical_loss": 3.965523213421738, "tokens_seen": 444268544 }, { "epoch": 0.13, "learning_rate": 0.00043705115346038116, "loss": 2.9746, "theoretical_loss": 3.965458924618173, "tokens_seen": 444334080 }, { "epoch": 0.13, "learning_rate": 0.0004370411233701103, "loss": 2.9064, "theoretical_loss": 3.965394647950581, "tokens_seen": 444399616 }, { "epoch": 0.13, "learning_rate": 0.0004370310932798395, "loss": 3.0837, "theoretical_loss": 3.9653303834148828, "tokens_seen": 444465152 }, { "epoch": 0.13, "learning_rate": 0.0004370210631895687, "loss": 2.8178, "theoretical_loss": 3.9652661310070005, "tokens_seen": 444530688 }, { "epoch": 0.13, "learning_rate": 0.0004370110330992979, "loss": 2.8228, "theoretical_loss": 3.965201890722857, "tokens_seen": 444596224 }, { "epoch": 0.13, "learning_rate": 0.00043700100300902707, "loss": 2.8377, "theoretical_loss": 3.9651376625583783, "tokens_seen": 444661760 }, { "epoch": 0.13, "learning_rate": 0.0004369909729187563, "loss": 3.2179, "theoretical_loss": 3.965073446509493, "tokens_seen": 444727296 }, { "epoch": 0.13, "learning_rate": 0.00043698094282848543, "loss": 2.8271, "theoretical_loss": 3.9650092425721297, "tokens_seen": 444792832 }, { "epoch": 0.13, "learning_rate": 0.00043697091273821467, "loss": 2.9445, "theoretical_loss": 3.9649450507422204, "tokens_seen": 444858368 }, { "epoch": 0.13, "learning_rate": 0.0004369608826479438, "loss": 3.0335, "theoretical_loss": 3.9648808710156986, "tokens_seen": 444923904 }, { "epoch": 0.13, "learning_rate": 0.00043695085255767303, "loss": 2.7964, "theoretical_loss": 3.964816703388501, "tokens_seen": 444989440 }, { "epoch": 0.13, "learning_rate": 0.0004369408224674022, "loss": 2.9693, "theoretical_loss": 3.9647525478565644, "tokens_seen": 445054976 }, { "epoch": 0.13, "learning_rate": 0.0004369307923771314, "loss": 2.8213, "theoretical_loss": 3.9646884044158277, "tokens_seen": 445120512 }, { "epoch": 0.13, "learning_rate": 0.0004369207622868606, "loss": 2.8553, "theoretical_loss": 3.964624273062234, "tokens_seen": 445186048 }, { "epoch": 0.13, "learning_rate": 0.00043691073219658975, "loss": 3.0907, "theoretical_loss": 3.9645601537917248, "tokens_seen": 445251584 }, { "epoch": 0.13, "learning_rate": 0.000436900702106319, "loss": 3.0419, "theoretical_loss": 3.964496046600247, "tokens_seen": 445317120 }, { "epoch": 0.13, "learning_rate": 0.00043689067201604817, "loss": 2.8665, "theoretical_loss": 3.9644319514837467, "tokens_seen": 445382656 }, { "epoch": 0.13, "learning_rate": 0.00043688064192577735, "loss": 3.0254, "theoretical_loss": 3.964367868438175, "tokens_seen": 445448192 }, { "epoch": 0.14, "learning_rate": 0.00043687061183550653, "loss": 2.9733, "theoretical_loss": 3.9643037974594817, "tokens_seen": 445513728 }, { "epoch": 0.14, "learning_rate": 0.00043686058174523577, "loss": 3.0529, "theoretical_loss": 3.96423973854362, "tokens_seen": 445579264 }, { "epoch": 0.14, "objective/train/docs_used": 264540, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0680365562438965, "objective/train/theoretical_loss": 3.9641756916865463, "objective/train/tokens_used": 466104800, "theoretical_loss": 3.9641756916865463, "tokens_seen": 445644800 }, { "epoch": 0.14, "learning_rate": 0.0004368505516549649, "loss": 2.7958, "theoretical_loss": 3.9641756916865463, "tokens_seen": 445644800 }, { "epoch": 0.14, "learning_rate": 0.00043684052156469413, "loss": 2.7212, "theoretical_loss": 3.964111656884217, "tokens_seen": 445710336 }, { "epoch": 0.14, "learning_rate": 0.00043683049147442326, "loss": 2.8679, "theoretical_loss": 3.9640476341325916, "tokens_seen": 445775872 }, { "epoch": 0.14, "learning_rate": 0.0004368204613841525, "loss": 2.8905, "theoretical_loss": 3.96398362342763, "tokens_seen": 445841408 }, { "epoch": 0.14, "learning_rate": 0.00043681043129388167, "loss": 3.0276, "theoretical_loss": 3.9639196247652966, "tokens_seen": 445906944 }, { "epoch": 0.14, "learning_rate": 0.00043680040120361085, "loss": 2.7716, "theoretical_loss": 3.9638556381415553, "tokens_seen": 445972480 }, { "epoch": 0.14, "learning_rate": 0.00043679037111334003, "loss": 2.97, "theoretical_loss": 3.9637916635523736, "tokens_seen": 446038016 }, { "epoch": 0.14, "learning_rate": 0.0004367803410230692, "loss": 3.0207, "theoretical_loss": 3.9637277009937195, "tokens_seen": 446103552 }, { "epoch": 0.14, "learning_rate": 0.0004367703109327984, "loss": 3.0506, "theoretical_loss": 3.9636637504615653, "tokens_seen": 446169088 }, { "epoch": 0.14, "learning_rate": 0.00043676028084252763, "loss": 2.8433, "theoretical_loss": 3.963599811951882, "tokens_seen": 446234624 }, { "epoch": 0.14, "learning_rate": 0.00043675025075225676, "loss": 2.9764, "theoretical_loss": 3.9635358854606455, "tokens_seen": 446300160 }, { "epoch": 0.14, "learning_rate": 0.000436740220661986, "loss": 2.8531, "theoretical_loss": 3.9634719709838317, "tokens_seen": 446365696 }, { "epoch": 0.14, "learning_rate": 0.0004367301905717151, "loss": 2.985, "theoretical_loss": 3.963408068517419, "tokens_seen": 446431232 }, { "epoch": 0.14, "learning_rate": 0.00043672016048144436, "loss": 2.9554, "theoretical_loss": 3.9633441780573886, "tokens_seen": 446496768 }, { "epoch": 0.14, "learning_rate": 0.00043671013039117354, "loss": 2.8556, "theoretical_loss": 3.963280299599722, "tokens_seen": 446562304 }, { "epoch": 0.14, "learning_rate": 0.0004367001003009027, "loss": 2.864, "theoretical_loss": 3.9632164331404045, "tokens_seen": 446627840 }, { "epoch": 0.14, "learning_rate": 0.0004366900702106319, "loss": 2.7874, "theoretical_loss": 3.9631525786754214, "tokens_seen": 446693376 }, { "epoch": 0.14, "learning_rate": 0.00043668004012036113, "loss": 2.7899, "theoretical_loss": 3.9630887362007616, "tokens_seen": 446758912 }, { "epoch": 0.14, "learning_rate": 0.00043667001003009026, "loss": 3.0638, "theoretical_loss": 3.9630249057124147, "tokens_seen": 446824448 }, { "epoch": 0.14, "learning_rate": 0.0004366599799398195, "loss": 2.8952, "theoretical_loss": 3.9629610872063727, "tokens_seen": 446889984 }, { "epoch": 0.14, "learning_rate": 0.0004366499498495486, "loss": 2.8926, "theoretical_loss": 3.9628972806786305, "tokens_seen": 446955520 }, { "epoch": 0.14, "learning_rate": 0.00043663991975927786, "loss": 2.74, "theoretical_loss": 3.9628334861251826, "tokens_seen": 447021056 }, { "epoch": 0.14, "learning_rate": 0.00043662988966900704, "loss": 2.9397, "theoretical_loss": 3.9627697035420284, "tokens_seen": 447086592 }, { "epoch": 0.14, "learning_rate": 0.0004366198595787362, "loss": 3.0488, "theoretical_loss": 3.9627059329251666, "tokens_seen": 447152128 }, { "epoch": 0.14, "learning_rate": 0.0004366098294884654, "loss": 2.946, "theoretical_loss": 3.9626421742705986, "tokens_seen": 447217664 }, { "epoch": 0.14, "objective/train/docs_used": 265849, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.205967664718628, "objective/train/theoretical_loss": 3.9625784275743294, "objective/train/tokens_used": 467743200, "theoretical_loss": 3.9625784275743294, "tokens_seen": 447283200 }, { "epoch": 0.14, "learning_rate": 0.0004365997993981946, "loss": 3.0183, "theoretical_loss": 3.9625784275743294, "tokens_seen": 447283200 }, { "epoch": 0.14, "learning_rate": 0.00043658976930792376, "loss": 2.8804, "theoretical_loss": 3.9625146928323627, "tokens_seen": 447348736 }, { "epoch": 0.14, "learning_rate": 0.000436579739217653, "loss": 2.8588, "theoretical_loss": 3.962450970040707, "tokens_seen": 447414272 }, { "epoch": 0.14, "learning_rate": 0.00043656970912738213, "loss": 2.9412, "theoretical_loss": 3.9623872591953715, "tokens_seen": 447479808 }, { "epoch": 0.14, "learning_rate": 0.00043655967903711136, "loss": 2.982, "theoretical_loss": 3.9623235602923677, "tokens_seen": 447545344 }, { "epoch": 0.14, "learning_rate": 0.0004365496489468405, "loss": 2.8512, "theoretical_loss": 3.9622598733277083, "tokens_seen": 447610880 }, { "epoch": 0.14, "learning_rate": 0.0004365396188565697, "loss": 2.707, "theoretical_loss": 3.9621961982974088, "tokens_seen": 447676416 }, { "epoch": 0.14, "learning_rate": 0.0004365295887662989, "loss": 3.0508, "theoretical_loss": 3.962132535197486, "tokens_seen": 447741952 }, { "epoch": 0.14, "learning_rate": 0.0004365195586760281, "loss": 2.8024, "theoretical_loss": 3.9620688840239593, "tokens_seen": 447807488 }, { "epoch": 0.14, "learning_rate": 0.00043650952858575727, "loss": 3.1377, "theoretical_loss": 3.9620052447728478, "tokens_seen": 447873024 }, { "epoch": 0.14, "learning_rate": 0.0004364994984954865, "loss": 2.8089, "theoretical_loss": 3.9619416174401767, "tokens_seen": 447938560 }, { "epoch": 0.14, "learning_rate": 0.00043648946840521563, "loss": 3.0379, "theoretical_loss": 3.9618780020219693, "tokens_seen": 448004096 }, { "epoch": 0.14, "learning_rate": 0.00043647943831494487, "loss": 2.8921, "theoretical_loss": 3.961814398514252, "tokens_seen": 448069632 }, { "epoch": 0.14, "learning_rate": 0.000436469408224674, "loss": 2.8712, "theoretical_loss": 3.961750806913054, "tokens_seen": 448135168 }, { "epoch": 0.14, "learning_rate": 0.00043645937813440323, "loss": 2.8566, "theoretical_loss": 3.961687227214406, "tokens_seen": 448200704 }, { "epoch": 0.14, "learning_rate": 0.0004364493480441324, "loss": 2.8192, "theoretical_loss": 3.961623659414338, "tokens_seen": 448266240 }, { "epoch": 0.14, "learning_rate": 0.0004364393179538616, "loss": 2.774, "theoretical_loss": 3.961560103508887, "tokens_seen": 448331776 }, { "epoch": 0.14, "learning_rate": 0.00043642928786359077, "loss": 3.0392, "theoretical_loss": 3.961496559494087, "tokens_seen": 448397312 }, { "epoch": 0.14, "learning_rate": 0.00043641925777331995, "loss": 2.9419, "theoretical_loss": 3.9614330273659775, "tokens_seen": 448462848 }, { "epoch": 0.14, "learning_rate": 0.00043640922768304913, "loss": 2.841, "theoretical_loss": 3.9613695071205974, "tokens_seen": 448528384 }, { "epoch": 0.14, "learning_rate": 0.00043639919759277837, "loss": 2.9117, "theoretical_loss": 3.9613059987539887, "tokens_seen": 448593920 }, { "epoch": 0.14, "learning_rate": 0.0004363891675025075, "loss": 2.7283, "theoretical_loss": 3.9612425022621953, "tokens_seen": 448659456 }, { "epoch": 0.14, "learning_rate": 0.00043637913741223673, "loss": 3.0715, "theoretical_loss": 3.9611790176412622, "tokens_seen": 448724992 }, { "epoch": 0.14, "learning_rate": 0.0004363691073219659, "loss": 2.7813, "theoretical_loss": 3.961115544887238, "tokens_seen": 448790528 }, { "epoch": 0.14, "learning_rate": 0.0004363590772316951, "loss": 2.8004, "theoretical_loss": 3.961052083996171, "tokens_seen": 448856064 }, { "epoch": 0.14, "objective/train/docs_used": 266141, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7323062419891357, "objective/train/theoretical_loss": 3.960988634964113, "objective/train/tokens_used": 469381600, "theoretical_loss": 3.960988634964113, "tokens_seen": 448921600 }, { "epoch": 0.14, "learning_rate": 0.0004363490471414243, "loss": 2.9443, "theoretical_loss": 3.960988634964113, "tokens_seen": 448921600 }, { "epoch": 0.14, "learning_rate": 0.00043633901705115346, "loss": 2.8353, "theoretical_loss": 3.960925197787116, "tokens_seen": 448987136 }, { "epoch": 0.14, "learning_rate": 0.00043632898696088264, "loss": 2.7991, "theoretical_loss": 3.9608617724612367, "tokens_seen": 449052672 }, { "epoch": 0.14, "learning_rate": 0.00043631895687061187, "loss": 3.0411, "theoretical_loss": 3.960798358982531, "tokens_seen": 449118208 }, { "epoch": 0.14, "learning_rate": 0.000436308926780341, "loss": 2.8481, "theoretical_loss": 3.960734957347058, "tokens_seen": 449183744 }, { "epoch": 0.14, "learning_rate": 0.00043629889669007023, "loss": 2.9892, "theoretical_loss": 3.960671567550878, "tokens_seen": 449249280 }, { "epoch": 0.14, "learning_rate": 0.00043628886659979936, "loss": 3.0781, "theoretical_loss": 3.9606081895900545, "tokens_seen": 449314816 }, { "epoch": 0.14, "learning_rate": 0.0004362788365095286, "loss": 2.7987, "theoretical_loss": 3.9605448234606513, "tokens_seen": 449380352 }, { "epoch": 0.14, "learning_rate": 0.0004362688064192578, "loss": 3.1004, "theoretical_loss": 3.9604814691587347, "tokens_seen": 449445888 }, { "epoch": 0.14, "learning_rate": 0.00043625877632898696, "loss": 2.9069, "theoretical_loss": 3.9604181266803726, "tokens_seen": 449511424 }, { "epoch": 0.14, "learning_rate": 0.00043624874623871614, "loss": 3.0532, "theoretical_loss": 3.960354796021636, "tokens_seen": 449576960 }, { "epoch": 0.14, "learning_rate": 0.0004362387161484453, "loss": 2.8056, "theoretical_loss": 3.960291477178596, "tokens_seen": 449642496 }, { "epoch": 0.14, "learning_rate": 0.0004362286860581745, "loss": 2.8333, "theoretical_loss": 3.960228170147327, "tokens_seen": 449708032 }, { "epoch": 0.14, "learning_rate": 0.00043621865596790374, "loss": 3.1691, "theoretical_loss": 3.9601648749239047, "tokens_seen": 449773568 }, { "epoch": 0.14, "learning_rate": 0.00043620862587763286, "loss": 2.9981, "theoretical_loss": 3.960101591504406, "tokens_seen": 449839104 }, { "epoch": 0.14, "learning_rate": 0.0004361985957873621, "loss": 3.0074, "theoretical_loss": 3.960038319884912, "tokens_seen": 449904640 }, { "epoch": 0.14, "learning_rate": 0.0004361885656970913, "loss": 2.7211, "theoretical_loss": 3.9599750600615025, "tokens_seen": 449970176 }, { "epoch": 0.14, "learning_rate": 0.00043617853560682046, "loss": 2.7248, "theoretical_loss": 3.9599118120302608, "tokens_seen": 450035712 }, { "epoch": 0.14, "learning_rate": 0.0004361685055165497, "loss": 2.7913, "theoretical_loss": 3.959848575787273, "tokens_seen": 450101248 }, { "epoch": 0.14, "learning_rate": 0.0004361584754262788, "loss": 2.513, "theoretical_loss": 3.959785351328626, "tokens_seen": 450166784 }, { "epoch": 0.14, "learning_rate": 0.00043614844533600806, "loss": 3.0842, "theoretical_loss": 3.9597221386504073, "tokens_seen": 450232320 }, { "epoch": 0.14, "learning_rate": 0.00043613841524573724, "loss": 2.9115, "theoretical_loss": 3.9596589377487086, "tokens_seen": 450297856 }, { "epoch": 0.14, "learning_rate": 0.0004361283851554664, "loss": 2.9273, "theoretical_loss": 3.9595957486196225, "tokens_seen": 450363392 }, { "epoch": 0.14, "learning_rate": 0.0004361183550651956, "loss": 3.0515, "theoretical_loss": 3.9595325712592433, "tokens_seen": 450428928 }, { "epoch": 0.14, "learning_rate": 0.0004361083249749248, "loss": 3.0497, "theoretical_loss": 3.9594694056636675, "tokens_seen": 450494464 }, { "epoch": 0.14, "objective/train/docs_used": 266764, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.3204047679901123, "objective/train/theoretical_loss": 3.9594062518289923, "objective/train/tokens_used": 471020000, "theoretical_loss": 3.9594062518289923, "tokens_seen": 450560000 }, { "epoch": 0.14, "learning_rate": 0.00043609829488465397, "loss": 2.9596, "theoretical_loss": 3.9594062518289923, "tokens_seen": 450560000 }, { "epoch": 0.14, "learning_rate": 0.0004360882647943832, "loss": 2.7606, "theoretical_loss": 3.9593431097513188, "tokens_seen": 450625536 }, { "epoch": 0.14, "learning_rate": 0.00043607823470411233, "loss": 3.0245, "theoretical_loss": 3.9592799794267486, "tokens_seen": 450691072 }, { "epoch": 0.14, "learning_rate": 0.00043606820461384156, "loss": 2.9339, "theoretical_loss": 3.9592168608513854, "tokens_seen": 450756608 }, { "epoch": 0.14, "learning_rate": 0.0004360581745235707, "loss": 3.0752, "theoretical_loss": 3.959153754021334, "tokens_seen": 450822144 }, { "epoch": 0.14, "learning_rate": 0.0004360481444332999, "loss": 2.9678, "theoretical_loss": 3.9590906589327037, "tokens_seen": 450887680 }, { "epoch": 0.14, "learning_rate": 0.0004360381143430291, "loss": 2.8541, "theoretical_loss": 3.9590275755816022, "tokens_seen": 450953216 }, { "epoch": 0.14, "learning_rate": 0.0004360280842527583, "loss": 2.8563, "theoretical_loss": 3.958964503964141, "tokens_seen": 451018752 }, { "epoch": 0.14, "learning_rate": 0.00043601805416248747, "loss": 3.1018, "theoretical_loss": 3.9589014440764334, "tokens_seen": 451084288 }, { "epoch": 0.14, "learning_rate": 0.0004360080240722167, "loss": 2.9896, "theoretical_loss": 3.9588383959145945, "tokens_seen": 451149824 }, { "epoch": 0.14, "learning_rate": 0.00043599799398194583, "loss": 2.8416, "theoretical_loss": 3.9587753594747404, "tokens_seen": 451215360 }, { "epoch": 0.14, "learning_rate": 0.00043598796389167507, "loss": 2.883, "theoretical_loss": 3.95871233475299, "tokens_seen": 451280896 }, { "epoch": 0.14, "learning_rate": 0.0004359779338014042, "loss": 2.9736, "theoretical_loss": 3.9586493217454635, "tokens_seen": 451346432 }, { "epoch": 0.14, "learning_rate": 0.00043596790371113343, "loss": 2.8513, "theoretical_loss": 3.9585863204482834, "tokens_seen": 451411968 }, { "epoch": 0.14, "learning_rate": 0.0004359578736208626, "loss": 3.0644, "theoretical_loss": 3.958523330857574, "tokens_seen": 451477504 }, { "epoch": 0.14, "learning_rate": 0.0004359478435305918, "loss": 2.7629, "theoretical_loss": 3.9584603529694604, "tokens_seen": 451543040 }, { "epoch": 0.14, "learning_rate": 0.00043593781344032097, "loss": 3.0606, "theoretical_loss": 3.9583973867800717, "tokens_seen": 451608576 }, { "epoch": 0.14, "learning_rate": 0.00043592778335005015, "loss": 2.8071, "theoretical_loss": 3.958334432285537, "tokens_seen": 451674112 }, { "epoch": 0.14, "learning_rate": 0.00043591775325977933, "loss": 2.9459, "theoretical_loss": 3.9582714894819873, "tokens_seen": 451739648 }, { "epoch": 0.14, "learning_rate": 0.00043590772316950857, "loss": 2.9942, "theoretical_loss": 3.9582085583655564, "tokens_seen": 451805184 }, { "epoch": 0.14, "learning_rate": 0.0004358976930792377, "loss": 3.0168, "theoretical_loss": 3.958145638932379, "tokens_seen": 451870720 }, { "epoch": 0.14, "learning_rate": 0.00043588766298896693, "loss": 3.036, "theoretical_loss": 3.9580827311785924, "tokens_seen": 451936256 }, { "epoch": 0.14, "learning_rate": 0.0004358776328986961, "loss": 2.8268, "theoretical_loss": 3.9580198351003357, "tokens_seen": 452001792 }, { "epoch": 0.14, "learning_rate": 0.0004358676028084253, "loss": 2.9403, "theoretical_loss": 3.9579569506937498, "tokens_seen": 452067328 }, { "epoch": 0.14, "learning_rate": 0.0004358575727181545, "loss": 2.7007, "theoretical_loss": 3.9578940779549763, "tokens_seen": 452132864 }, { "epoch": 0.14, "objective/train/docs_used": 268192, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8176727294921875, "objective/train/theoretical_loss": 3.9578312168801597, "objective/train/tokens_used": 472658400, "theoretical_loss": 3.9578312168801597, "tokens_seen": 452198400 }, { "epoch": 0.14, "learning_rate": 0.00043584754262788366, "loss": 2.9893, "theoretical_loss": 3.9578312168801597, "tokens_seen": 452198400 }, { "epoch": 0.14, "learning_rate": 0.00043583751253761284, "loss": 3.0281, "theoretical_loss": 3.957768367465447, "tokens_seen": 452263936 }, { "epoch": 0.14, "learning_rate": 0.00043582748244734207, "loss": 2.9604, "theoretical_loss": 3.9577055297069856, "tokens_seen": 452329472 }, { "epoch": 0.14, "learning_rate": 0.0004358174523570712, "loss": 2.8216, "theoretical_loss": 3.957642703600926, "tokens_seen": 452395008 }, { "epoch": 0.14, "learning_rate": 0.00043580742226680043, "loss": 3.2024, "theoretical_loss": 3.9575798891434184, "tokens_seen": 452460544 }, { "epoch": 0.14, "learning_rate": 0.00043579739217652956, "loss": 2.9174, "theoretical_loss": 3.957517086330618, "tokens_seen": 452526080 }, { "epoch": 0.14, "learning_rate": 0.0004357873620862588, "loss": 3.0686, "theoretical_loss": 3.957454295158679, "tokens_seen": 452591616 }, { "epoch": 0.14, "learning_rate": 0.000435777331995988, "loss": 2.946, "theoretical_loss": 3.9573915156237587, "tokens_seen": 452657152 }, { "epoch": 0.14, "learning_rate": 0.00043576730190571716, "loss": 3.0834, "theoretical_loss": 3.9573287477220167, "tokens_seen": 452722688 }, { "epoch": 0.14, "learning_rate": 0.00043575727181544634, "loss": 3.2719, "theoretical_loss": 3.957265991449613, "tokens_seen": 452788224 }, { "epoch": 0.14, "learning_rate": 0.0004357472417251755, "loss": 2.8808, "theoretical_loss": 3.957203246802711, "tokens_seen": 452853760 }, { "epoch": 0.14, "learning_rate": 0.0004357372116349047, "loss": 3.0469, "theoretical_loss": 3.9571405137774747, "tokens_seen": 452919296 }, { "epoch": 0.14, "learning_rate": 0.00043572718154463394, "loss": 2.9176, "theoretical_loss": 3.9570777923700704, "tokens_seen": 452984832 }, { "epoch": 0.14, "learning_rate": 0.00043571715145436306, "loss": 2.864, "theoretical_loss": 3.9570150825766657, "tokens_seen": 453050368 }, { "epoch": 0.14, "learning_rate": 0.0004357071213640923, "loss": 3.0054, "theoretical_loss": 3.956952384393432, "tokens_seen": 453115904 }, { "epoch": 0.14, "learning_rate": 0.0004356970912738215, "loss": 2.9216, "theoretical_loss": 3.956889697816539, "tokens_seen": 453181440 }, { "epoch": 0.14, "learning_rate": 0.00043568706118355066, "loss": 3.0549, "theoretical_loss": 3.9568270228421616, "tokens_seen": 453246976 }, { "epoch": 0.14, "learning_rate": 0.00043567703109327984, "loss": 2.995, "theoretical_loss": 3.956764359466475, "tokens_seen": 453312512 }, { "epoch": 0.14, "learning_rate": 0.000435667001003009, "loss": 2.7352, "theoretical_loss": 3.9567017076856565, "tokens_seen": 453378048 }, { "epoch": 0.14, "learning_rate": 0.0004356569709127382, "loss": 2.9331, "theoretical_loss": 3.9566390674958845, "tokens_seen": 453443584 }, { "epoch": 0.14, "learning_rate": 0.00043564694082246744, "loss": 3.1616, "theoretical_loss": 3.9565764388933395, "tokens_seen": 453509120 }, { "epoch": 0.14, "learning_rate": 0.00043563691073219657, "loss": 2.943, "theoretical_loss": 3.9565138218742053, "tokens_seen": 453574656 }, { "epoch": 0.14, "learning_rate": 0.0004356268806419258, "loss": 2.9295, "theoretical_loss": 3.9564512164346652, "tokens_seen": 453640192 }, { "epoch": 0.14, "learning_rate": 0.00043561685055165493, "loss": 3.067, "theoretical_loss": 3.956388622570906, "tokens_seen": 453705728 }, { "epoch": 0.14, "learning_rate": 0.00043560682046138417, "loss": 3.1012, "theoretical_loss": 3.956326040279116, "tokens_seen": 453771264 }, { "epoch": 0.14, "objective/train/docs_used": 268717, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.054826021194458, "objective/train/theoretical_loss": 3.956263469555484, "objective/train/tokens_used": 474296800, "theoretical_loss": 3.956263469555484, "tokens_seen": 453836800 }, { "epoch": 0.14, "learning_rate": 0.00043559679037111335, "loss": 2.8934, "theoretical_loss": 3.956263469555484, "tokens_seen": 453836800 }, { "epoch": 0.14, "learning_rate": 0.00043558676028084253, "loss": 3.0294, "theoretical_loss": 3.956200910396202, "tokens_seen": 453902336 }, { "epoch": 0.14, "learning_rate": 0.0004355767301905717, "loss": 2.7316, "theoretical_loss": 3.9561383627974642, "tokens_seen": 453967872 }, { "epoch": 0.14, "learning_rate": 0.0004355667001003009, "loss": 2.82, "theoretical_loss": 3.956075826755465, "tokens_seen": 454033408 }, { "epoch": 0.14, "learning_rate": 0.00043555667001003007, "loss": 2.59, "theoretical_loss": 3.9560133022664026, "tokens_seen": 454098944 }, { "epoch": 0.14, "learning_rate": 0.0004355466399197593, "loss": 2.8896, "theoretical_loss": 3.955950789326474, "tokens_seen": 454164480 }, { "epoch": 0.14, "learning_rate": 0.00043553660982948843, "loss": 2.7608, "theoretical_loss": 3.955888287931881, "tokens_seen": 454230016 }, { "epoch": 0.14, "learning_rate": 0.00043552657973921767, "loss": 2.8049, "theoretical_loss": 3.955825798078826, "tokens_seen": 454295552 }, { "epoch": 0.14, "learning_rate": 0.00043551654964894685, "loss": 2.7974, "theoretical_loss": 3.9557633197635127, "tokens_seen": 454361088 }, { "epoch": 0.14, "learning_rate": 0.00043550651955867603, "loss": 2.8448, "theoretical_loss": 3.9557008529821474, "tokens_seen": 454426624 }, { "epoch": 0.14, "learning_rate": 0.0004354964894684052, "loss": 2.9746, "theoretical_loss": 3.955638397730939, "tokens_seen": 454492160 }, { "epoch": 0.14, "learning_rate": 0.0004354864593781344, "loss": 2.9851, "theoretical_loss": 3.955575954006095, "tokens_seen": 454557696 }, { "epoch": 0.14, "learning_rate": 0.0004354764292878636, "loss": 3.1936, "theoretical_loss": 3.9555135218038284, "tokens_seen": 454623232 }, { "epoch": 0.14, "learning_rate": 0.0004354663991975928, "loss": 2.8998, "theoretical_loss": 3.955451101120352, "tokens_seen": 454688768 }, { "epoch": 0.14, "learning_rate": 0.00043545636910732194, "loss": 2.8088, "theoretical_loss": 3.955388691951881, "tokens_seen": 454754304 }, { "epoch": 0.14, "learning_rate": 0.00043544633901705117, "loss": 2.9158, "theoretical_loss": 3.9553262942946312, "tokens_seen": 454819840 }, { "epoch": 0.14, "learning_rate": 0.0004354363089267803, "loss": 3.0216, "theoretical_loss": 3.9552639081448224, "tokens_seen": 454885376 }, { "epoch": 0.14, "learning_rate": 0.00043542627883650953, "loss": 2.8275, "theoretical_loss": 3.9552015334986743, "tokens_seen": 454950912 }, { "epoch": 0.14, "learning_rate": 0.00043541624874623877, "loss": 2.887, "theoretical_loss": 3.9551391703524086, "tokens_seen": 455016448 }, { "epoch": 0.14, "learning_rate": 0.0004354062186559679, "loss": 2.9557, "theoretical_loss": 3.95507681870225, "tokens_seen": 455081984 }, { "epoch": 0.14, "learning_rate": 0.00043539618856569713, "loss": 2.8015, "theoretical_loss": 3.955014478544425, "tokens_seen": 455147520 }, { "epoch": 0.14, "learning_rate": 0.0004353861584754263, "loss": 3.0108, "theoretical_loss": 3.954952149875159, "tokens_seen": 455213056 }, { "epoch": 0.14, "learning_rate": 0.0004353761283851555, "loss": 2.8855, "theoretical_loss": 3.9548898326906823, "tokens_seen": 455278592 }, { "epoch": 0.14, "learning_rate": 0.0004353660982948847, "loss": 3.0832, "theoretical_loss": 3.954827526987226, "tokens_seen": 455344128 }, { "epoch": 0.14, "learning_rate": 0.00043535606820461386, "loss": 3.0242, "theoretical_loss": 3.954765232761023, "tokens_seen": 455409664 }, { "epoch": 0.14, "objective/train/docs_used": 269967, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.981569290161133, "objective/train/theoretical_loss": 3.954702950008308, "objective/train/tokens_used": 475935200, "theoretical_loss": 3.954702950008308, "tokens_seen": 455475200 }, { "epoch": 0.14, "learning_rate": 0.00043534603811434304, "loss": 3.0046, "theoretical_loss": 3.954702950008308, "tokens_seen": 455475200 }, { "epoch": 0.14, "learning_rate": 0.00043533600802407227, "loss": 3.0107, "theoretical_loss": 3.954640678725317, "tokens_seen": 455540736 }, { "epoch": 0.14, "learning_rate": 0.0004353259779338014, "loss": 2.8819, "theoretical_loss": 3.9545784189082887, "tokens_seen": 455606272 }, { "epoch": 0.14, "learning_rate": 0.00043531594784353063, "loss": 2.7767, "theoretical_loss": 3.9545161705534624, "tokens_seen": 455671808 }, { "epoch": 0.14, "learning_rate": 0.00043530591775325976, "loss": 3.0823, "theoretical_loss": 3.9544539336570796, "tokens_seen": 455737344 }, { "epoch": 0.14, "learning_rate": 0.000435295887662989, "loss": 2.8641, "theoretical_loss": 3.954391708215385, "tokens_seen": 455802880 }, { "epoch": 0.14, "learning_rate": 0.0004352858575727182, "loss": 2.9376, "theoretical_loss": 3.954329494224623, "tokens_seen": 455868416 }, { "epoch": 0.14, "learning_rate": 0.00043527582748244736, "loss": 2.9881, "theoretical_loss": 3.9542672916810404, "tokens_seen": 455933952 }, { "epoch": 0.14, "learning_rate": 0.00043526579739217654, "loss": 2.596, "theoretical_loss": 3.9542051005808867, "tokens_seen": 455999488 }, { "epoch": 0.14, "learning_rate": 0.0004352557673019057, "loss": 2.9525, "theoretical_loss": 3.954142920920412, "tokens_seen": 456065024 }, { "epoch": 0.14, "learning_rate": 0.0004352457372116349, "loss": 3.0159, "theoretical_loss": 3.954080752695868, "tokens_seen": 456130560 }, { "epoch": 0.14, "learning_rate": 0.00043523570712136414, "loss": 3.0215, "theoretical_loss": 3.95401859590351, "tokens_seen": 456196096 }, { "epoch": 0.14, "learning_rate": 0.00043522567703109326, "loss": 2.8253, "theoretical_loss": 3.953956450539594, "tokens_seen": 456261632 }, { "epoch": 0.14, "learning_rate": 0.0004352156469408225, "loss": 2.9501, "theoretical_loss": 3.953894316600376, "tokens_seen": 456327168 }, { "epoch": 0.14, "learning_rate": 0.0004352056168505517, "loss": 2.9843, "theoretical_loss": 3.953832194082117, "tokens_seen": 456392704 }, { "epoch": 0.14, "learning_rate": 0.00043519558676028086, "loss": 3.1067, "theoretical_loss": 3.9537700829810767, "tokens_seen": 456458240 }, { "epoch": 0.14, "learning_rate": 0.00043518555667001004, "loss": 2.8276, "theoretical_loss": 3.95370798329352, "tokens_seen": 456523776 }, { "epoch": 0.14, "learning_rate": 0.0004351755265797392, "loss": 2.8316, "theoretical_loss": 3.9536458950157094, "tokens_seen": 456589312 }, { "epoch": 0.14, "learning_rate": 0.0004351654964894684, "loss": 2.6923, "theoretical_loss": 3.9535838181439127, "tokens_seen": 456654848 }, { "epoch": 0.14, "learning_rate": 0.00043515546639919764, "loss": 2.7719, "theoretical_loss": 3.9535217526743978, "tokens_seen": 456720384 }, { "epoch": 0.14, "learning_rate": 0.00043514543630892677, "loss": 2.9918, "theoretical_loss": 3.953459698603434, "tokens_seen": 456785920 }, { "epoch": 0.14, "learning_rate": 0.000435135406218656, "loss": 2.9902, "theoretical_loss": 3.9533976559272936, "tokens_seen": 456851456 }, { "epoch": 0.14, "learning_rate": 0.00043512537612838513, "loss": 2.9201, "theoretical_loss": 3.95333562464225, "tokens_seen": 456916992 }, { "epoch": 0.14, "learning_rate": 0.00043511534603811437, "loss": 3.0096, "theoretical_loss": 3.9532736047445787, "tokens_seen": 456982528 }, { "epoch": 0.14, "learning_rate": 0.00043510531594784355, "loss": 2.8607, "theoretical_loss": 3.953211596230556, "tokens_seen": 457048064 }, { "epoch": 0.14, "objective/train/docs_used": 270622, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.566558361053467, "objective/train/theoretical_loss": 3.953149599096461, "objective/train/tokens_used": 477573600, "theoretical_loss": 3.953149599096461, "tokens_seen": 457113600 }, { "epoch": 0.14, "learning_rate": 0.00043509528585757273, "loss": 2.7782, "theoretical_loss": 3.953149599096461, "tokens_seen": 457113600 }, { "epoch": 0.14, "learning_rate": 0.0004350852557673019, "loss": 2.9516, "theoretical_loss": 3.9530876133385737, "tokens_seen": 457179136 }, { "epoch": 0.14, "learning_rate": 0.0004350752256770311, "loss": 3.0632, "theoretical_loss": 3.953025638953177, "tokens_seen": 457244672 }, { "epoch": 0.14, "learning_rate": 0.00043506519558676027, "loss": 3.1882, "theoretical_loss": 3.9529636759365543, "tokens_seen": 457310208 }, { "epoch": 0.14, "learning_rate": 0.0004350551654964895, "loss": 2.9347, "theoretical_loss": 3.9529017242849918, "tokens_seen": 457375744 }, { "epoch": 0.14, "learning_rate": 0.00043504513540621863, "loss": 2.8628, "theoretical_loss": 3.952839783994776, "tokens_seen": 457441280 }, { "epoch": 0.14, "learning_rate": 0.00043503510531594787, "loss": 2.875, "theoretical_loss": 3.9527778550621973, "tokens_seen": 457506816 }, { "epoch": 0.14, "learning_rate": 0.00043502507522567705, "loss": 2.9098, "theoretical_loss": 3.9527159374835463, "tokens_seen": 457572352 }, { "epoch": 0.14, "learning_rate": 0.00043501504513540623, "loss": 2.9462, "theoretical_loss": 3.952654031255115, "tokens_seen": 457637888 }, { "epoch": 0.14, "learning_rate": 0.0004350050150451354, "loss": 3.0357, "theoretical_loss": 3.9525921363731977, "tokens_seen": 457703424 }, { "epoch": 0.14, "learning_rate": 0.0004349949849548646, "loss": 2.8875, "theoretical_loss": 3.9525302528340918, "tokens_seen": 457768960 }, { "epoch": 0.14, "learning_rate": 0.0004349849548645938, "loss": 2.8557, "theoretical_loss": 3.9524683806340937, "tokens_seen": 457834496 }, { "epoch": 0.14, "learning_rate": 0.000434974924774323, "loss": 2.9504, "theoretical_loss": 3.9524065197695046, "tokens_seen": 457900032 }, { "epoch": 0.14, "learning_rate": 0.00043496489468405214, "loss": 3.0495, "theoretical_loss": 3.9523446702366245, "tokens_seen": 457965568 }, { "epoch": 0.14, "learning_rate": 0.00043495486459378137, "loss": 2.9949, "theoretical_loss": 3.9522828320317576, "tokens_seen": 458031104 }, { "epoch": 0.14, "learning_rate": 0.0004349448345035105, "loss": 3.1083, "theoretical_loss": 3.9522210051512077, "tokens_seen": 458096640 }, { "epoch": 0.14, "learning_rate": 0.00043493480441323973, "loss": 2.9009, "theoretical_loss": 3.9521591895912818, "tokens_seen": 458162176 }, { "epoch": 0.14, "learning_rate": 0.0004349247743229689, "loss": 2.9672, "theoretical_loss": 3.9520973853482886, "tokens_seen": 458227712 }, { "epoch": 0.14, "learning_rate": 0.0004349147442326981, "loss": 2.7928, "theoretical_loss": 3.9520355924185377, "tokens_seen": 458293248 }, { "epoch": 0.14, "learning_rate": 0.0004349047141424273, "loss": 2.789, "theoretical_loss": 3.951973810798341, "tokens_seen": 458358784 }, { "epoch": 0.14, "learning_rate": 0.0004348946840521565, "loss": 2.956, "theoretical_loss": 3.951912040484012, "tokens_seen": 458424320 }, { "epoch": 0.14, "learning_rate": 0.00043488465396188564, "loss": 2.9218, "theoretical_loss": 3.9518502814718652, "tokens_seen": 458489856 }, { "epoch": 0.14, "learning_rate": 0.0004348746238716149, "loss": 3.0027, "theoretical_loss": 3.9517885337582186, "tokens_seen": 458555392 }, { "epoch": 0.14, "learning_rate": 0.000434864593781344, "loss": 2.722, "theoretical_loss": 3.951726797339391, "tokens_seen": 458620928 }, { "epoch": 0.14, "learning_rate": 0.00043485456369107324, "loss": 2.9446, "theoretical_loss": 3.9516650722117017, "tokens_seen": 458686464 }, { "epoch": 0.14, "objective/train/docs_used": 271952, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9281325340270996, "objective/train/theoretical_loss": 3.9516033583714734, "objective/train/tokens_used": 479212000, "theoretical_loss": 3.9516033583714734, "tokens_seen": 458752000 }, { "epoch": 0.14, "learning_rate": 0.0004348445336008024, "loss": 2.9981, "theoretical_loss": 3.9516033583714734, "tokens_seen": 458752000 }, { "epoch": 0.14, "learning_rate": 0.0004348345035105316, "loss": 2.7936, "theoretical_loss": 3.9515416558150305, "tokens_seen": 458817536 }, { "epoch": 0.14, "learning_rate": 0.0004348244734202608, "loss": 2.8275, "theoretical_loss": 3.9514799645386973, "tokens_seen": 458883072 }, { "epoch": 0.14, "learning_rate": 0.00043481444332998996, "loss": 2.9014, "theoretical_loss": 3.9514182845388026, "tokens_seen": 458948608 }, { "epoch": 0.14, "learning_rate": 0.00043480441323971914, "loss": 2.8708, "theoretical_loss": 3.951356615811674, "tokens_seen": 459014144 }, { "epoch": 0.14, "learning_rate": 0.0004347943831494484, "loss": 3.1129, "theoretical_loss": 3.9512949583536434, "tokens_seen": 459079680 }, { "epoch": 0.14, "learning_rate": 0.0004347843530591775, "loss": 2.603, "theoretical_loss": 3.951233312161042, "tokens_seen": 459145216 }, { "epoch": 0.14, "learning_rate": 0.00043477432296890674, "loss": 2.888, "theoretical_loss": 3.9511716772302057, "tokens_seen": 459210752 }, { "epoch": 0.14, "learning_rate": 0.00043476429287863587, "loss": 2.6878, "theoretical_loss": 3.9511100535574686, "tokens_seen": 459276288 }, { "epoch": 0.14, "learning_rate": 0.0004347542627883651, "loss": 2.896, "theoretical_loss": 3.9510484411391698, "tokens_seen": 459341824 }, { "epoch": 0.14, "learning_rate": 0.0004347442326980943, "loss": 2.7514, "theoretical_loss": 3.9509868399716472, "tokens_seen": 459407360 }, { "epoch": 0.14, "learning_rate": 0.00043473420260782346, "loss": 2.6764, "theoretical_loss": 3.9509252500512426, "tokens_seen": 459472896 }, { "epoch": 0.14, "learning_rate": 0.00043472417251755265, "loss": 2.9212, "theoretical_loss": 3.950863671374299, "tokens_seen": 459538432 }, { "epoch": 0.14, "learning_rate": 0.0004347141424272819, "loss": 3.0134, "theoretical_loss": 3.9508021039371597, "tokens_seen": 459603968 }, { "epoch": 0.14, "learning_rate": 0.000434704112337011, "loss": 2.9894, "theoretical_loss": 3.9507405477361726, "tokens_seen": 459669504 }, { "epoch": 0.14, "learning_rate": 0.00043469408224674024, "loss": 3.1493, "theoretical_loss": 3.9506790027676844, "tokens_seen": 459735040 }, { "epoch": 0.14, "learning_rate": 0.0004346840521564694, "loss": 2.9438, "theoretical_loss": 3.9506174690280442, "tokens_seen": 459800576 }, { "epoch": 0.14, "learning_rate": 0.0004346740220661986, "loss": 3.0193, "theoretical_loss": 3.950555946513605, "tokens_seen": 459866112 }, { "epoch": 0.14, "learning_rate": 0.00043466399197592784, "loss": 3.1924, "theoretical_loss": 3.9504944352207176, "tokens_seen": 459931648 }, { "epoch": 0.14, "learning_rate": 0.00043465396188565697, "loss": 2.9121, "theoretical_loss": 3.9504329351457383, "tokens_seen": 459997184 }, { "epoch": 0.14, "learning_rate": 0.0004346439317953862, "loss": 2.8726, "theoretical_loss": 3.9503714462850237, "tokens_seen": 460062720 }, { "epoch": 0.14, "learning_rate": 0.00043463390170511533, "loss": 2.7697, "theoretical_loss": 3.9503099686349303, "tokens_seen": 460128256 }, { "epoch": 0.14, "learning_rate": 0.00043462387161484457, "loss": 2.9298, "theoretical_loss": 3.950248502191819, "tokens_seen": 460193792 }, { "epoch": 0.14, "learning_rate": 0.00043461384152457375, "loss": 3.136, "theoretical_loss": 3.950187046952051, "tokens_seen": 460259328 }, { "epoch": 0.14, "learning_rate": 0.00043460381143430293, "loss": 2.7644, "theoretical_loss": 3.9501256029119896, "tokens_seen": 460324864 }, { "epoch": 0.14, "objective/train/docs_used": 272724, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.293632984161377, "objective/train/theoretical_loss": 3.950064170068, "objective/train/tokens_used": 480850400, "theoretical_loss": 3.950064170068, "tokens_seen": 460390400 }, { "epoch": 0.14, "learning_rate": 0.0004345937813440321, "loss": 2.7199, "theoretical_loss": 3.950064170068, "tokens_seen": 460390400 }, { "epoch": 0.14, "learning_rate": 0.0004345837512537613, "loss": 3.0363, "theoretical_loss": 3.9500027484164484, "tokens_seen": 460455936 }, { "epoch": 0.14, "learning_rate": 0.00043457372116349047, "loss": 2.9059, "theoretical_loss": 3.9499413379537027, "tokens_seen": 460521472 }, { "epoch": 0.14, "learning_rate": 0.0004345636910732197, "loss": 2.9963, "theoretical_loss": 3.9498799386761334, "tokens_seen": 460587008 }, { "epoch": 0.14, "learning_rate": 0.00043455366098294883, "loss": 2.8352, "theoretical_loss": 3.949818550580112, "tokens_seen": 460652544 }, { "epoch": 0.14, "learning_rate": 0.00043454363089267807, "loss": 3.0915, "theoretical_loss": 3.9497571736620127, "tokens_seen": 460718080 }, { "epoch": 0.14, "learning_rate": 0.00043453360080240725, "loss": 2.9781, "theoretical_loss": 3.949695807918209, "tokens_seen": 460783616 }, { "epoch": 0.14, "learning_rate": 0.00043452357071213643, "loss": 3.1897, "theoretical_loss": 3.9496344533450793, "tokens_seen": 460849152 }, { "epoch": 0.14, "learning_rate": 0.0004345135406218656, "loss": 2.7796, "theoretical_loss": 3.949573109939001, "tokens_seen": 460914688 }, { "epoch": 0.14, "learning_rate": 0.0004345035105315948, "loss": 2.9629, "theoretical_loss": 3.9495117776963538, "tokens_seen": 460980224 }, { "epoch": 0.14, "learning_rate": 0.000434493480441324, "loss": 2.961, "theoretical_loss": 3.949450456613521, "tokens_seen": 461045760 }, { "epoch": 0.14, "learning_rate": 0.0004344834503510532, "loss": 2.9782, "theoretical_loss": 3.9493891466868845, "tokens_seen": 461111296 }, { "epoch": 0.14, "learning_rate": 0.00043447342026078234, "loss": 3.1149, "theoretical_loss": 3.949327847912831, "tokens_seen": 461176832 }, { "epoch": 0.14, "learning_rate": 0.00043446339017051157, "loss": 2.9404, "theoretical_loss": 3.9492665602877457, "tokens_seen": 461242368 }, { "epoch": 0.14, "learning_rate": 0.0004344533600802407, "loss": 2.8125, "theoretical_loss": 3.9492052838080185, "tokens_seen": 461307904 }, { "epoch": 0.14, "learning_rate": 0.00043444332998996993, "loss": 2.6222, "theoretical_loss": 3.9491440184700393, "tokens_seen": 461373440 }, { "epoch": 0.14, "learning_rate": 0.0004344332998996991, "loss": 2.9793, "theoretical_loss": 3.9490827642701998, "tokens_seen": 461438976 }, { "epoch": 0.14, "learning_rate": 0.0004344232698094283, "loss": 2.938, "theoretical_loss": 3.9490215212048936, "tokens_seen": 461504512 }, { "epoch": 0.14, "learning_rate": 0.0004344132397191575, "loss": 2.6944, "theoretical_loss": 3.9489602892705165, "tokens_seen": 461570048 }, { "epoch": 0.14, "learning_rate": 0.0004344032096288867, "loss": 2.7612, "theoretical_loss": 3.948899068463465, "tokens_seen": 461635584 }, { "epoch": 0.14, "learning_rate": 0.00043439317953861584, "loss": 2.8936, "theoretical_loss": 3.948837858780138, "tokens_seen": 461701120 }, { "epoch": 0.14, "learning_rate": 0.0004343831494483451, "loss": 2.9781, "theoretical_loss": 3.948776660216935, "tokens_seen": 461766656 }, { "epoch": 0.14, "learning_rate": 0.0004343731193580742, "loss": 3.1315, "theoretical_loss": 3.948715472770259, "tokens_seen": 461832192 }, { "epoch": 0.14, "learning_rate": 0.00043436308926780344, "loss": 2.9559, "theoretical_loss": 3.948654296436513, "tokens_seen": 461897728 }, { "epoch": 0.14, "learning_rate": 0.0004343530591775326, "loss": 3.0797, "theoretical_loss": 3.9485931312121028, "tokens_seen": 461963264 }, { "epoch": 0.14, "objective/train/docs_used": 273842, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.113386869430542, "objective/train/theoretical_loss": 3.9485319770934355, "objective/train/tokens_used": 482488800, "theoretical_loss": 3.9485319770934355, "tokens_seen": 462028800 }, { "epoch": 0.14, "learning_rate": 0.0004343430290872618, "loss": 2.9541, "theoretical_loss": 3.9485319770934355, "tokens_seen": 462028800 }, { "epoch": 0.14, "learning_rate": 0.000434332998996991, "loss": 3.0204, "theoretical_loss": 3.948470834076919, "tokens_seen": 462094336 }, { "epoch": 0.14, "learning_rate": 0.00043432296890672016, "loss": 2.8271, "theoretical_loss": 3.9484097021589646, "tokens_seen": 462159872 }, { "epoch": 0.14, "learning_rate": 0.00043431293881644934, "loss": 3.0506, "theoretical_loss": 3.948348581335984, "tokens_seen": 462225408 }, { "epoch": 0.14, "learning_rate": 0.0004343029087261786, "loss": 2.939, "theoretical_loss": 3.94828747160439, "tokens_seen": 462290944 }, { "epoch": 0.14, "learning_rate": 0.0004342928786359077, "loss": 2.7207, "theoretical_loss": 3.9482263729605993, "tokens_seen": 462356480 }, { "epoch": 0.14, "learning_rate": 0.00043428284854563694, "loss": 2.8159, "theoretical_loss": 3.9481652854010285, "tokens_seen": 462422016 }, { "epoch": 0.14, "learning_rate": 0.00043427281845536607, "loss": 2.9092, "theoretical_loss": 3.948104208922096, "tokens_seen": 462487552 }, { "epoch": 0.14, "learning_rate": 0.0004342627883650953, "loss": 2.7242, "theoretical_loss": 3.9480431435202226, "tokens_seen": 462553088 }, { "epoch": 0.14, "learning_rate": 0.0004342527582748245, "loss": 2.8413, "theoretical_loss": 3.947982089191829, "tokens_seen": 462618624 }, { "epoch": 0.14, "learning_rate": 0.00043424272818455367, "loss": 2.6571, "theoretical_loss": 3.9479210459333407, "tokens_seen": 462684160 }, { "epoch": 0.14, "learning_rate": 0.00043423269809428285, "loss": 2.7971, "theoretical_loss": 3.9478600137411823, "tokens_seen": 462749696 }, { "epoch": 0.14, "learning_rate": 0.0004342226680040121, "loss": 2.6765, "theoretical_loss": 3.94779899261178, "tokens_seen": 462815232 }, { "epoch": 0.14, "learning_rate": 0.0004342126379137412, "loss": 2.9225, "theoretical_loss": 3.9477379825415637, "tokens_seen": 462880768 }, { "epoch": 0.14, "learning_rate": 0.00043420260782347044, "loss": 3.0055, "theoretical_loss": 3.9476769835269634, "tokens_seen": 462946304 }, { "epoch": 0.14, "learning_rate": 0.00043419257773319957, "loss": 2.8627, "theoretical_loss": 3.9476159955644112, "tokens_seen": 463011840 }, { "epoch": 0.14, "learning_rate": 0.0004341825476429288, "loss": 2.9815, "theoretical_loss": 3.94755501865034, "tokens_seen": 463077376 }, { "epoch": 0.14, "learning_rate": 0.000434172517552658, "loss": 2.7327, "theoretical_loss": 3.947494052781185, "tokens_seen": 463142912 }, { "epoch": 0.14, "learning_rate": 0.00043416248746238717, "loss": 2.8695, "theoretical_loss": 3.9474330979533843, "tokens_seen": 463208448 }, { "epoch": 0.14, "learning_rate": 0.00043415245737211635, "loss": 3.2073, "theoretical_loss": 3.947372154163375, "tokens_seen": 463273984 }, { "epoch": 0.14, "learning_rate": 0.00043414242728184553, "loss": 2.9631, "theoretical_loss": 3.947311221407599, "tokens_seen": 463339520 }, { "epoch": 0.14, "learning_rate": 0.0004341323971915747, "loss": 2.9367, "theoretical_loss": 3.947250299682498, "tokens_seen": 463405056 }, { "epoch": 0.14, "learning_rate": 0.00043412236710130395, "loss": 2.9153, "theoretical_loss": 3.9471893889845138, "tokens_seen": 463470592 }, { "epoch": 0.14, "learning_rate": 0.0004341123370110331, "loss": 2.9749, "theoretical_loss": 3.947128489310093, "tokens_seen": 463536128 }, { "epoch": 0.14, "learning_rate": 0.0004341023069207623, "loss": 2.8678, "theoretical_loss": 3.947067600655682, "tokens_seen": 463601664 }, { "epoch": 0.14, "objective/train/docs_used": 274430, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4908626079559326, "objective/train/theoretical_loss": 3.94700672301773, "objective/train/tokens_used": 484127200, "theoretical_loss": 3.94700672301773, "tokens_seen": 463667200 }, { "epoch": 0.14, "learning_rate": 0.00043409227683049144, "loss": 2.5694, "theoretical_loss": 3.94700672301773, "tokens_seen": 463667200 }, { "epoch": 0.14, "learning_rate": 0.00043408224674022067, "loss": 2.717, "theoretical_loss": 3.946945856392687, "tokens_seen": 463732736 }, { "epoch": 0.14, "learning_rate": 0.00043407221664994985, "loss": 2.6902, "theoretical_loss": 3.9468850007770033, "tokens_seen": 463798272 }, { "epoch": 0.14, "learning_rate": 0.00043406218655967903, "loss": 2.786, "theoretical_loss": 3.9468241561671342, "tokens_seen": 463863808 }, { "epoch": 0.14, "learning_rate": 0.0004340521564694082, "loss": 2.4293, "theoretical_loss": 3.946763322559534, "tokens_seen": 463929344 }, { "epoch": 0.14, "learning_rate": 0.00043404212637913745, "loss": 2.6931, "theoretical_loss": 3.946702499950659, "tokens_seen": 463994880 }, { "epoch": 0.14, "learning_rate": 0.0004340320962888666, "loss": 3.0037, "theoretical_loss": 3.9466416883369684, "tokens_seen": 464060416 }, { "epoch": 0.14, "learning_rate": 0.0004340220661985958, "loss": 2.9052, "theoretical_loss": 3.946580887714921, "tokens_seen": 464125952 }, { "epoch": 0.14, "learning_rate": 0.00043401203610832494, "loss": 2.8656, "theoretical_loss": 3.9465200980809803, "tokens_seen": 464191488 }, { "epoch": 0.14, "learning_rate": 0.0004340020060180542, "loss": 3.08, "theoretical_loss": 3.946459319431608, "tokens_seen": 464257024 }, { "epoch": 0.14, "learning_rate": 0.00043399197592778336, "loss": 2.8027, "theoretical_loss": 3.9463985517632696, "tokens_seen": 464322560 }, { "epoch": 0.14, "learning_rate": 0.00043398194583751254, "loss": 2.8701, "theoretical_loss": 3.9463377950724317, "tokens_seen": 464388096 }, { "epoch": 0.14, "learning_rate": 0.0004339719157472417, "loss": 2.9089, "theoretical_loss": 3.9462770493555617, "tokens_seen": 464453632 }, { "epoch": 0.14, "learning_rate": 0.0004339618856569709, "loss": 2.8239, "theoretical_loss": 3.9462163146091305, "tokens_seen": 464519168 }, { "epoch": 0.14, "learning_rate": 0.0004339518555667001, "loss": 2.8597, "theoretical_loss": 3.9461555908296093, "tokens_seen": 464584704 }, { "epoch": 0.14, "learning_rate": 0.0004339418254764293, "loss": 2.8746, "theoretical_loss": 3.9460948780134713, "tokens_seen": 464650240 }, { "epoch": 0.14, "learning_rate": 0.0004339317953861585, "loss": 2.9924, "theoretical_loss": 3.94603417615719, "tokens_seen": 464715776 }, { "epoch": 0.14, "learning_rate": 0.0004339217652958877, "loss": 2.8558, "theoretical_loss": 3.9459734852572437, "tokens_seen": 464781312 }, { "epoch": 0.14, "learning_rate": 0.0004339117352056169, "loss": 3.0925, "theoretical_loss": 3.9459128053101082, "tokens_seen": 464846848 }, { "epoch": 0.14, "learning_rate": 0.00043390170511534604, "loss": 2.9914, "theoretical_loss": 3.945852136312265, "tokens_seen": 464912384 }, { "epoch": 0.14, "learning_rate": 0.0004338916750250753, "loss": 3.0237, "theoretical_loss": 3.945791478260195, "tokens_seen": 464977920 }, { "epoch": 0.14, "learning_rate": 0.0004338816449348044, "loss": 2.5918, "theoretical_loss": 3.94573083115038, "tokens_seen": 465043456 }, { "epoch": 0.14, "learning_rate": 0.00043387161484453364, "loss": 2.6599, "theoretical_loss": 3.9456701949793054, "tokens_seen": 465108992 }, { "epoch": 0.14, "learning_rate": 0.0004338615847542628, "loss": 3.0173, "theoretical_loss": 3.9456095697434566, "tokens_seen": 465174528 }, { "epoch": 0.14, "learning_rate": 0.000433851554663992, "loss": 2.9841, "theoretical_loss": 3.945548955439322, "tokens_seen": 465240064 }, { "epoch": 0.14, "objective/train/docs_used": 275809, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.787060260772705, "objective/train/theoretical_loss": 3.945488352063391, "objective/train/tokens_used": 485765600, "theoretical_loss": 3.945488352063391, "tokens_seen": 465305600 }, { "epoch": 0.14, "learning_rate": 0.0004338415245737212, "loss": 2.9794, "theoretical_loss": 3.945488352063391, "tokens_seen": 465305600 }, { "epoch": 0.14, "learning_rate": 0.00043383149448345036, "loss": 2.8995, "theoretical_loss": 3.945427759612154, "tokens_seen": 465371136 }, { "epoch": 0.14, "learning_rate": 0.00043382146439317954, "loss": 2.6856, "theoretical_loss": 3.945367178082104, "tokens_seen": 465436672 }, { "epoch": 0.14, "learning_rate": 0.0004338114343029088, "loss": 2.8025, "theoretical_loss": 3.945306607469735, "tokens_seen": 465502208 }, { "epoch": 0.14, "learning_rate": 0.0004338014042126379, "loss": 2.8469, "theoretical_loss": 3.945246047771543, "tokens_seen": 465567744 }, { "epoch": 0.14, "learning_rate": 0.00043379137412236714, "loss": 3.0447, "theoretical_loss": 3.9451854989840256, "tokens_seen": 465633280 }, { "epoch": 0.14, "learning_rate": 0.00043378134403209627, "loss": 2.7582, "theoretical_loss": 3.945124961103682, "tokens_seen": 465698816 }, { "epoch": 0.14, "learning_rate": 0.0004337713139418255, "loss": 2.9184, "theoretical_loss": 3.9450644341270116, "tokens_seen": 465764352 }, { "epoch": 0.14, "learning_rate": 0.0004337612838515547, "loss": 2.9787, "theoretical_loss": 3.945003918050518, "tokens_seen": 465829888 }, { "epoch": 0.14, "learning_rate": 0.00043375125376128387, "loss": 2.8831, "theoretical_loss": 3.9449434128707055, "tokens_seen": 465895424 }, { "epoch": 0.14, "learning_rate": 0.00043374122367101305, "loss": 2.9134, "theoretical_loss": 3.9448829185840784, "tokens_seen": 465960960 }, { "epoch": 0.14, "learning_rate": 0.0004337311935807423, "loss": 2.7942, "theoretical_loss": 3.9448224351871444, "tokens_seen": 466026496 }, { "epoch": 0.14, "learning_rate": 0.0004337211634904714, "loss": 3.0798, "theoretical_loss": 3.944761962676412, "tokens_seen": 466092032 }, { "epoch": 0.14, "learning_rate": 0.00043371113340020064, "loss": 3.0969, "theoretical_loss": 3.9447015010483923, "tokens_seen": 466157568 }, { "epoch": 0.14, "learning_rate": 0.00043370110330992977, "loss": 2.7817, "theoretical_loss": 3.9446410502995963, "tokens_seen": 466223104 }, { "epoch": 0.14, "learning_rate": 0.000433691073219659, "loss": 2.8557, "theoretical_loss": 3.9445806104265384, "tokens_seen": 466288640 }, { "epoch": 0.14, "learning_rate": 0.0004336810431293882, "loss": 2.8694, "theoretical_loss": 3.9445201814257334, "tokens_seen": 466354176 }, { "epoch": 0.14, "learning_rate": 0.00043367101303911737, "loss": 2.973, "theoretical_loss": 3.9444597632936977, "tokens_seen": 466419712 }, { "epoch": 0.14, "learning_rate": 0.00043366098294884655, "loss": 2.6732, "theoretical_loss": 3.9443993560269504, "tokens_seen": 466485248 }, { "epoch": 0.14, "learning_rate": 0.00043365095285857573, "loss": 2.9185, "theoretical_loss": 3.9443389596220113, "tokens_seen": 466550784 }, { "epoch": 0.14, "learning_rate": 0.0004336409227683049, "loss": 2.7752, "theoretical_loss": 3.944278574075402, "tokens_seen": 466616320 }, { "epoch": 0.14, "learning_rate": 0.00043363089267803415, "loss": 2.8808, "theoretical_loss": 3.944218199383645, "tokens_seen": 466681856 }, { "epoch": 0.14, "learning_rate": 0.0004336208625877633, "loss": 2.8209, "theoretical_loss": 3.9441578355432663, "tokens_seen": 466747392 }, { "epoch": 0.14, "learning_rate": 0.0004336108324974925, "loss": 2.7177, "theoretical_loss": 3.9440974825507915, "tokens_seen": 466812928 }, { "epoch": 0.14, "learning_rate": 0.00043360080240722164, "loss": 2.7596, "theoretical_loss": 3.9440371404027488, "tokens_seen": 466878464 }, { "epoch": 0.14, "objective/train/docs_used": 276320, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.076183795928955, "objective/train/theoretical_loss": 3.943976809095668, "objective/train/tokens_used": 487404000, "theoretical_loss": 3.943976809095668, "tokens_seen": 466944000 }, { "epoch": 0.14, "learning_rate": 0.00043359077231695087, "loss": 3.1843, "theoretical_loss": 3.943976809095668, "tokens_seen": 466944000 }, { "epoch": 0.14, "learning_rate": 0.00043358074222668005, "loss": 2.9046, "theoretical_loss": 3.94391648862608, "tokens_seen": 467009536 }, { "epoch": 0.14, "learning_rate": 0.00043357071213640923, "loss": 2.9376, "theoretical_loss": 3.9438561789905178, "tokens_seen": 467075072 }, { "epoch": 0.14, "learning_rate": 0.0004335606820461384, "loss": 2.887, "theoretical_loss": 3.9437958801855157, "tokens_seen": 467140608 }, { "epoch": 0.14, "learning_rate": 0.00043355065195586765, "loss": 2.7006, "theoretical_loss": 3.9437355922076094, "tokens_seen": 467206144 }, { "epoch": 0.14, "learning_rate": 0.0004335406218655968, "loss": 2.6553, "theoretical_loss": 3.9436753150533375, "tokens_seen": 467271680 }, { "epoch": 0.14, "learning_rate": 0.000433530591775326, "loss": 2.8684, "theoretical_loss": 3.943615048719238, "tokens_seen": 467337216 }, { "epoch": 0.14, "learning_rate": 0.00043352056168505514, "loss": 2.8836, "theoretical_loss": 3.9435547932018524, "tokens_seen": 467402752 }, { "epoch": 0.14, "learning_rate": 0.0004335105315947844, "loss": 2.9624, "theoretical_loss": 3.943494548497723, "tokens_seen": 467468288 }, { "epoch": 0.14, "learning_rate": 0.00043350050150451356, "loss": 2.6615, "theoretical_loss": 3.943434314603394, "tokens_seen": 467533824 }, { "epoch": 0.14, "learning_rate": 0.00043349047141424274, "loss": 3.0893, "theoretical_loss": 3.9433740915154094, "tokens_seen": 467599360 }, { "epoch": 0.14, "learning_rate": 0.0004334804413239719, "loss": 3.1646, "theoretical_loss": 3.9433138792303177, "tokens_seen": 467664896 }, { "epoch": 0.14, "learning_rate": 0.0004334704112337011, "loss": 3.1173, "theoretical_loss": 3.943253677744668, "tokens_seen": 467730432 }, { "epoch": 0.14, "learning_rate": 0.0004334603811434303, "loss": 2.7868, "theoretical_loss": 3.943193487055009, "tokens_seen": 467795968 }, { "epoch": 0.14, "learning_rate": 0.0004334503510531595, "loss": 2.7069, "theoretical_loss": 3.9431333071578942, "tokens_seen": 467861504 }, { "epoch": 0.14, "learning_rate": 0.00043344032096288864, "loss": 2.8466, "theoretical_loss": 3.9430731380498765, "tokens_seen": 467927040 }, { "epoch": 0.14, "learning_rate": 0.0004334302908726179, "loss": 2.9696, "theoretical_loss": 3.94301297972751, "tokens_seen": 467992576 }, { "epoch": 0.14, "learning_rate": 0.000433420260782347, "loss": 2.6652, "theoretical_loss": 3.9429528321873537, "tokens_seen": 468058112 }, { "epoch": 0.14, "learning_rate": 0.00043341023069207624, "loss": 2.9861, "theoretical_loss": 3.942892695425963, "tokens_seen": 468123648 }, { "epoch": 0.14, "learning_rate": 0.0004334002006018054, "loss": 2.9029, "theoretical_loss": 3.9428325694398993, "tokens_seen": 468189184 }, { "epoch": 0.14, "learning_rate": 0.0004333901705115346, "loss": 2.8075, "theoretical_loss": 3.942772454225724, "tokens_seen": 468254720 }, { "epoch": 0.14, "learning_rate": 0.0004333801404212638, "loss": 2.8315, "theoretical_loss": 3.94271234978, "tokens_seen": 468320256 }, { "epoch": 0.14, "learning_rate": 0.000433370110330993, "loss": 3.0407, "theoretical_loss": 3.9426522560992905, "tokens_seen": 468385792 }, { "epoch": 0.14, "learning_rate": 0.00043336008024072215, "loss": 2.7421, "theoretical_loss": 3.942592173180163, "tokens_seen": 468451328 }, { "epoch": 0.14, "learning_rate": 0.0004333500501504514, "loss": 2.9915, "theoretical_loss": 3.9425321010191854, "tokens_seen": 468516864 }, { "epoch": 0.14, "objective/train/docs_used": 277290, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1820871829986572, "objective/train/theoretical_loss": 3.942472039612926, "objective/train/tokens_used": 489042400, "theoretical_loss": 3.942472039612926, "tokens_seen": 468582400 }, { "epoch": 0.14, "learning_rate": 0.0004333400200601805, "loss": 2.9808, "theoretical_loss": 3.942472039612926, "tokens_seen": 468582400 }, { "epoch": 0.14, "learning_rate": 0.00043332998996990974, "loss": 3.1072, "theoretical_loss": 3.9424119889579563, "tokens_seen": 468647936 }, { "epoch": 0.14, "learning_rate": 0.0004333199598796389, "loss": 2.8486, "theoretical_loss": 3.9423519490508476, "tokens_seen": 468713472 }, { "epoch": 0.14, "learning_rate": 0.0004333099297893681, "loss": 2.8733, "theoretical_loss": 3.9422919198881754, "tokens_seen": 468779008 }, { "epoch": 0.14, "learning_rate": 0.0004332998996990973, "loss": 2.9926, "theoretical_loss": 3.9422319014665144, "tokens_seen": 468844544 }, { "epoch": 0.14, "learning_rate": 0.00043328986960882647, "loss": 2.9607, "theoretical_loss": 3.942171893782442, "tokens_seen": 468910080 }, { "epoch": 0.14, "learning_rate": 0.00043327983951855565, "loss": 2.7794, "theoretical_loss": 3.942111896832537, "tokens_seen": 468975616 }, { "epoch": 0.14, "learning_rate": 0.0004332698094282849, "loss": 2.9306, "theoretical_loss": 3.9420519106133787, "tokens_seen": 469041152 }, { "epoch": 0.14, "learning_rate": 0.000433259779338014, "loss": 3.2266, "theoretical_loss": 3.9419919351215498, "tokens_seen": 469106688 }, { "epoch": 0.14, "learning_rate": 0.00043324974924774325, "loss": 2.9204, "theoretical_loss": 3.9419319703536337, "tokens_seen": 469172224 }, { "epoch": 0.14, "learning_rate": 0.00043323971915747243, "loss": 2.9933, "theoretical_loss": 3.941872016306215, "tokens_seen": 469237760 }, { "epoch": 0.14, "learning_rate": 0.0004332296890672016, "loss": 2.8745, "theoretical_loss": 3.941812072975881, "tokens_seen": 469303296 }, { "epoch": 0.14, "learning_rate": 0.0004332196589769308, "loss": 2.9306, "theoretical_loss": 3.9417521403592186, "tokens_seen": 469368832 }, { "epoch": 0.14, "learning_rate": 0.00043320962888665997, "loss": 2.6249, "theoretical_loss": 3.9416922184528183, "tokens_seen": 469434368 }, { "epoch": 0.14, "learning_rate": 0.00043319959879638915, "loss": 2.9339, "theoretical_loss": 3.94163230725327, "tokens_seen": 469499904 }, { "epoch": 0.14, "learning_rate": 0.0004331895687061184, "loss": 2.9625, "theoretical_loss": 3.9415724067571687, "tokens_seen": 469565440 }, { "epoch": 0.14, "learning_rate": 0.00043317953861584757, "loss": 3.0605, "theoretical_loss": 3.941512516961107, "tokens_seen": 469630976 }, { "epoch": 0.14, "learning_rate": 0.00043316950852557675, "loss": 2.9603, "theoretical_loss": 3.941452637861681, "tokens_seen": 469696512 }, { "epoch": 0.14, "learning_rate": 0.00043315947843530593, "loss": 2.817, "theoretical_loss": 3.9413927694554887, "tokens_seen": 469762048 }, { "epoch": 0.14, "learning_rate": 0.0004331494483450351, "loss": 2.9923, "theoretical_loss": 3.9413329117391287, "tokens_seen": 469827584 }, { "epoch": 0.14, "learning_rate": 0.00043313941825476435, "loss": 2.8578, "theoretical_loss": 3.9412730647092022, "tokens_seen": 469893120 }, { "epoch": 0.14, "learning_rate": 0.0004331293881644935, "loss": 2.7678, "theoretical_loss": 3.94121322836231, "tokens_seen": 469958656 }, { "epoch": 0.14, "learning_rate": 0.0004331193580742227, "loss": 2.7757, "theoretical_loss": 3.9411534026950568, "tokens_seen": 470024192 }, { "epoch": 0.14, "learning_rate": 0.00043310932798395184, "loss": 2.8787, "theoretical_loss": 3.9410935877040476, "tokens_seen": 470089728 }, { "epoch": 0.14, "learning_rate": 0.00043309929789368107, "loss": 2.9, "theoretical_loss": 3.9410337833858886, "tokens_seen": 470155264 }, { "epoch": 0.14, "objective/train/docs_used": 277851, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8162131309509277, "objective/train/theoretical_loss": 3.9409739897371896, "objective/train/tokens_used": 490680800, "theoretical_loss": 3.9409739897371896, "tokens_seen": 470220800 }, { "epoch": 0.14, "learning_rate": 0.00043308926780341025, "loss": 3.1145, "theoretical_loss": 3.9409739897371896, "tokens_seen": 470220800 }, { "epoch": 0.14, "learning_rate": 0.00043307923771313943, "loss": 2.9894, "theoretical_loss": 3.9409142067545586, "tokens_seen": 470286336 }, { "epoch": 0.14, "learning_rate": 0.0004330692076228686, "loss": 3.0744, "theoretical_loss": 3.940854434434608, "tokens_seen": 470351872 }, { "epoch": 0.14, "learning_rate": 0.00043305917753259785, "loss": 2.8549, "theoretical_loss": 3.940794672773951, "tokens_seen": 470417408 }, { "epoch": 0.14, "learning_rate": 0.000433049147442327, "loss": 2.7925, "theoretical_loss": 3.9407349217692023, "tokens_seen": 470482944 }, { "epoch": 0.14, "learning_rate": 0.0004330391173520562, "loss": 2.9453, "theoretical_loss": 3.9406751814169763, "tokens_seen": 470548480 }, { "epoch": 0.14, "learning_rate": 0.00043302908726178534, "loss": 2.8738, "theoretical_loss": 3.9406154517138923, "tokens_seen": 470614016 }, { "epoch": 0.14, "learning_rate": 0.0004330190571715146, "loss": 2.7708, "theoretical_loss": 3.940555732656569, "tokens_seen": 470679552 }, { "epoch": 0.14, "learning_rate": 0.00043300902708124376, "loss": 3.0653, "theoretical_loss": 3.940496024241627, "tokens_seen": 470745088 }, { "epoch": 0.14, "learning_rate": 0.00043299899699097294, "loss": 3.0183, "theoretical_loss": 3.9404363264656883, "tokens_seen": 470810624 }, { "epoch": 0.14, "learning_rate": 0.0004329889669007021, "loss": 2.9802, "theoretical_loss": 3.9403766393253767, "tokens_seen": 470876160 }, { "epoch": 0.14, "learning_rate": 0.0004329789368104313, "loss": 2.7904, "theoretical_loss": 3.9403169628173185, "tokens_seen": 470941696 }, { "epoch": 0.14, "learning_rate": 0.0004329689067201605, "loss": 2.8852, "theoretical_loss": 3.9402572969381393, "tokens_seen": 471007232 }, { "epoch": 0.14, "learning_rate": 0.0004329588766298897, "loss": 2.9922, "theoretical_loss": 3.9401976416844677, "tokens_seen": 471072768 }, { "epoch": 0.14, "learning_rate": 0.00043294884653961884, "loss": 2.8197, "theoretical_loss": 3.9401379970529344, "tokens_seen": 471138304 }, { "epoch": 0.14, "learning_rate": 0.0004329388164493481, "loss": 3.0398, "theoretical_loss": 3.94007836304017, "tokens_seen": 471203840 }, { "epoch": 0.14, "learning_rate": 0.0004329287863590772, "loss": 3.0596, "theoretical_loss": 3.9400187396428086, "tokens_seen": 471269376 }, { "epoch": 0.14, "learning_rate": 0.00043291875626880644, "loss": 3.0314, "theoretical_loss": 3.9399591268574836, "tokens_seen": 471334912 }, { "epoch": 0.14, "learning_rate": 0.0004329087261785356, "loss": 3.0936, "theoretical_loss": 3.9398995246808317, "tokens_seen": 471400448 }, { "epoch": 0.14, "learning_rate": 0.0004328986960882648, "loss": 2.6104, "theoretical_loss": 3.93983993310949, "tokens_seen": 471465984 }, { "epoch": 0.14, "learning_rate": 0.000432888665997994, "loss": 2.8409, "theoretical_loss": 3.9397803521400983, "tokens_seen": 471531520 }, { "epoch": 0.14, "learning_rate": 0.0004328786359077232, "loss": 3.0642, "theoretical_loss": 3.9397207817692967, "tokens_seen": 471597056 }, { "epoch": 0.14, "learning_rate": 0.00043286860581745235, "loss": 2.7443, "theoretical_loss": 3.9396612219937284, "tokens_seen": 471662592 }, { "epoch": 0.14, "learning_rate": 0.0004328585757271816, "loss": 2.738, "theoretical_loss": 3.9396016728100363, "tokens_seen": 471728128 }, { "epoch": 0.14, "learning_rate": 0.0004328485456369107, "loss": 2.8538, "theoretical_loss": 3.9395421342148653, "tokens_seen": 471793664 }, { "epoch": 0.14, "objective/train/docs_used": 279270, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0141212940216064, "objective/train/theoretical_loss": 3.939482606204863, "objective/train/tokens_used": 492319200, "theoretical_loss": 3.939482606204863, "tokens_seen": 471859200 }, { "epoch": 0.14, "learning_rate": 0.00043283851554663994, "loss": 2.8997, "theoretical_loss": 3.939482606204863, "tokens_seen": 471859200 }, { "epoch": 0.14, "learning_rate": 0.0004328284854563691, "loss": 3.008, "theoretical_loss": 3.9394230887766772, "tokens_seen": 471924736 }, { "epoch": 0.14, "learning_rate": 0.0004328184553660983, "loss": 3.0273, "theoretical_loss": 3.939363581926959, "tokens_seen": 471990272 }, { "epoch": 0.14, "learning_rate": 0.0004328084252758275, "loss": 3.1478, "theoretical_loss": 3.939304085652358, "tokens_seen": 472055808 }, { "epoch": 0.14, "learning_rate": 0.00043279839518555667, "loss": 3.0269, "theoretical_loss": 3.939244599949528, "tokens_seen": 472121344 }, { "epoch": 0.14, "learning_rate": 0.00043278836509528585, "loss": 2.7709, "theoretical_loss": 3.939185124815123, "tokens_seen": 472186880 }, { "epoch": 0.14, "learning_rate": 0.0004327783350050151, "loss": 2.7383, "theoretical_loss": 3.9391256602457996, "tokens_seen": 472252416 }, { "epoch": 0.14, "learning_rate": 0.0004327683049147442, "loss": 2.7578, "theoretical_loss": 3.9390662062382145, "tokens_seen": 472317952 }, { "epoch": 0.14, "learning_rate": 0.00043275827482447345, "loss": 2.7545, "theoretical_loss": 3.9390067627890275, "tokens_seen": 472383488 }, { "epoch": 0.14, "learning_rate": 0.00043274824473420263, "loss": 2.9205, "theoretical_loss": 3.938947329894899, "tokens_seen": 472449024 }, { "epoch": 0.14, "learning_rate": 0.0004327382146439318, "loss": 3.0668, "theoretical_loss": 3.9388879075524907, "tokens_seen": 472514560 }, { "epoch": 0.14, "learning_rate": 0.000432728184553661, "loss": 2.7883, "theoretical_loss": 3.9388284957584663, "tokens_seen": 472580096 }, { "epoch": 0.14, "learning_rate": 0.00043271815446339017, "loss": 2.7036, "theoretical_loss": 3.9387690945094906, "tokens_seen": 472645632 }, { "epoch": 0.14, "learning_rate": 0.00043270812437311935, "loss": 3.0538, "theoretical_loss": 3.93870970380223, "tokens_seen": 472711168 }, { "epoch": 0.14, "learning_rate": 0.0004326980942828486, "loss": 2.8628, "theoretical_loss": 3.938650323633354, "tokens_seen": 472776704 }, { "epoch": 0.14, "learning_rate": 0.0004326880641925777, "loss": 2.9396, "theoretical_loss": 3.9385909539995305, "tokens_seen": 472842240 }, { "epoch": 0.14, "learning_rate": 0.00043267803410230695, "loss": 3.0433, "theoretical_loss": 3.9385315948974315, "tokens_seen": 472907776 }, { "epoch": 0.14, "learning_rate": 0.0004326680040120361, "loss": 2.772, "theoretical_loss": 3.9384722463237294, "tokens_seen": 472973312 }, { "epoch": 0.14, "learning_rate": 0.0004326579739217653, "loss": 2.9285, "theoretical_loss": 3.938412908275099, "tokens_seen": 473038848 }, { "epoch": 0.14, "learning_rate": 0.0004326479438314945, "loss": 2.8074, "theoretical_loss": 3.938353580748215, "tokens_seen": 473104384 }, { "epoch": 0.14, "learning_rate": 0.0004326379137412237, "loss": 2.6641, "theoretical_loss": 3.938294263739755, "tokens_seen": 473169920 }, { "epoch": 0.14, "learning_rate": 0.00043262788365095286, "loss": 2.8836, "theoretical_loss": 3.9382349572463973, "tokens_seen": 473235456 }, { "epoch": 0.14, "learning_rate": 0.00043261785356068204, "loss": 3.0981, "theoretical_loss": 3.938175661264823, "tokens_seen": 473300992 }, { "epoch": 0.14, "learning_rate": 0.0004326078234704112, "loss": 3.0504, "theoretical_loss": 3.9381163757917133, "tokens_seen": 473366528 }, { "epoch": 0.14, "learning_rate": 0.00043259779338014045, "loss": 3.0151, "theoretical_loss": 3.938057100823751, "tokens_seen": 473432064 }, { "epoch": 0.14, "objective/train/docs_used": 279967, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.561666250228882, "objective/train/theoretical_loss": 3.9379978363576216, "objective/train/tokens_used": 493957600, "theoretical_loss": 3.9379978363576216, "tokens_seen": 473497600 }, { "epoch": 0.14, "learning_rate": 0.0004325877632898696, "loss": 2.6292, "theoretical_loss": 3.9379978363576216, "tokens_seen": 473497600 }, { "epoch": 0.14, "learning_rate": 0.0004325777331995988, "loss": 3.0368, "theoretical_loss": 3.937938582390011, "tokens_seen": 473563136 }, { "epoch": 0.14, "learning_rate": 0.000432567703109328, "loss": 3.079, "theoretical_loss": 3.9378793389176066, "tokens_seen": 473628672 }, { "epoch": 0.14, "learning_rate": 0.0004325576730190572, "loss": 2.7806, "theoretical_loss": 3.9378201059370976, "tokens_seen": 473694208 }, { "epoch": 0.14, "learning_rate": 0.00043254764292878636, "loss": 2.6743, "theoretical_loss": 3.9377608834451756, "tokens_seen": 473759744 }, { "epoch": 0.14, "learning_rate": 0.00043253761283851554, "loss": 3.1192, "theoretical_loss": 3.9377016714385316, "tokens_seen": 473825280 }, { "epoch": 0.14, "learning_rate": 0.0004325275827482447, "loss": 2.9094, "theoretical_loss": 3.9376424699138606, "tokens_seen": 473890816 }, { "epoch": 0.14, "learning_rate": 0.00043251755265797396, "loss": 2.7598, "theoretical_loss": 3.937583278867857, "tokens_seen": 473956352 }, { "epoch": 0.14, "learning_rate": 0.0004325075225677031, "loss": 2.8362, "theoretical_loss": 3.9375240982972173, "tokens_seen": 474021888 }, { "epoch": 0.14, "learning_rate": 0.0004324974924774323, "loss": 3.0722, "theoretical_loss": 3.9374649281986405, "tokens_seen": 474087424 }, { "epoch": 0.14, "learning_rate": 0.00043248746238716145, "loss": 2.7437, "theoretical_loss": 3.937405768568826, "tokens_seen": 474152960 }, { "epoch": 0.14, "learning_rate": 0.0004324774322968907, "loss": 2.9541, "theoretical_loss": 3.937346619404475, "tokens_seen": 474218496 }, { "epoch": 0.14, "learning_rate": 0.00043246740220661986, "loss": 2.8422, "theoretical_loss": 3.93728748070229, "tokens_seen": 474284032 }, { "epoch": 0.14, "learning_rate": 0.00043245737211634904, "loss": 2.7602, "theoretical_loss": 3.937228352458976, "tokens_seen": 474349568 }, { "epoch": 0.14, "learning_rate": 0.0004324473420260782, "loss": 2.9715, "theoretical_loss": 3.9371692346712375, "tokens_seen": 474415104 }, { "epoch": 0.14, "learning_rate": 0.0004324373119358074, "loss": 3.0922, "theoretical_loss": 3.9371101273357825, "tokens_seen": 474480640 }, { "epoch": 0.14, "learning_rate": 0.00043242728184553664, "loss": 2.8695, "theoretical_loss": 3.93705103044932, "tokens_seen": 474546176 }, { "epoch": 0.14, "learning_rate": 0.0004324172517552658, "loss": 2.9216, "theoretical_loss": 3.9369919440085592, "tokens_seen": 474611712 }, { "epoch": 0.14, "learning_rate": 0.000432407221664995, "loss": 2.8716, "theoretical_loss": 3.9369328680102122, "tokens_seen": 474677248 }, { "epoch": 0.14, "learning_rate": 0.0004323971915747242, "loss": 2.9434, "theoretical_loss": 3.9368738024509926, "tokens_seen": 474742784 }, { "epoch": 0.14, "learning_rate": 0.0004323871614844534, "loss": 3.2621, "theoretical_loss": 3.9368147473276145, "tokens_seen": 474808320 }, { "epoch": 0.14, "learning_rate": 0.00043237713139418255, "loss": 2.9925, "theoretical_loss": 3.936755702636794, "tokens_seen": 474873856 }, { "epoch": 0.14, "learning_rate": 0.0004323671013039118, "loss": 2.744, "theoretical_loss": 3.9366966683752493, "tokens_seen": 474939392 }, { "epoch": 0.14, "learning_rate": 0.0004323570712136409, "loss": 2.7103, "theoretical_loss": 3.936637644539699, "tokens_seen": 475004928 }, { "epoch": 0.14, "learning_rate": 0.00043234704112337014, "loss": 2.7144, "theoretical_loss": 3.9365786311268636, "tokens_seen": 475070464 }, { "epoch": 0.14, "objective/train/docs_used": 280356, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8112435340881348, "objective/train/theoretical_loss": 3.936519628133466, "objective/train/tokens_used": 495596000, "theoretical_loss": 3.936519628133466, "tokens_seen": 475136000 }, { "epoch": 0.14, "learning_rate": 0.0004323370110330993, "loss": 3.1572, "theoretical_loss": 3.936519628133466, "tokens_seen": 475136000 }, { "epoch": 0.14, "learning_rate": 0.0004323269809428285, "loss": 2.7392, "theoretical_loss": 3.9364606355562284, "tokens_seen": 475201536 }, { "epoch": 0.14, "learning_rate": 0.0004323169508525577, "loss": 2.6772, "theoretical_loss": 3.9364016533918766, "tokens_seen": 475267072 }, { "epoch": 0.14, "learning_rate": 0.00043230692076228687, "loss": 3.0261, "theoretical_loss": 3.9363426816371376, "tokens_seen": 475332608 }, { "epoch": 0.14, "learning_rate": 0.00043229689067201605, "loss": 2.7741, "theoretical_loss": 3.9362837202887384, "tokens_seen": 475398144 }, { "epoch": 0.14, "learning_rate": 0.0004322868605817453, "loss": 2.9289, "theoretical_loss": 3.9362247693434096, "tokens_seen": 475463680 }, { "epoch": 0.14, "learning_rate": 0.0004322768304914744, "loss": 2.7103, "theoretical_loss": 3.9361658287978813, "tokens_seen": 475529216 }, { "epoch": 0.14, "learning_rate": 0.00043226680040120365, "loss": 2.6407, "theoretical_loss": 3.9361068986488856, "tokens_seen": 475594752 }, { "epoch": 0.14, "learning_rate": 0.00043225677031093283, "loss": 3.1158, "theoretical_loss": 3.936047978893158, "tokens_seen": 475660288 }, { "epoch": 0.14, "learning_rate": 0.000432246740220662, "loss": 2.9055, "theoretical_loss": 3.935989069527432, "tokens_seen": 475725824 }, { "epoch": 0.14, "learning_rate": 0.0004322367101303912, "loss": 2.8514, "theoretical_loss": 3.9359301705484455, "tokens_seen": 475791360 }, { "epoch": 0.14, "learning_rate": 0.00043222668004012037, "loss": 2.8603, "theoretical_loss": 3.935871281952937, "tokens_seen": 475856896 }, { "epoch": 0.14, "learning_rate": 0.00043221664994984955, "loss": 2.8935, "theoretical_loss": 3.935812403737646, "tokens_seen": 475922432 }, { "epoch": 0.14, "learning_rate": 0.0004322066198595788, "loss": 3.0784, "theoretical_loss": 3.9357535358993134, "tokens_seen": 475987968 }, { "epoch": 0.14, "learning_rate": 0.0004321965897693079, "loss": 2.7309, "theoretical_loss": 3.9356946784346825, "tokens_seen": 476053504 }, { "epoch": 0.14, "learning_rate": 0.00043218655967903715, "loss": 2.8205, "theoretical_loss": 3.9356358313404973, "tokens_seen": 476119040 }, { "epoch": 0.14, "learning_rate": 0.0004321765295887663, "loss": 2.8091, "theoretical_loss": 3.9355769946135037, "tokens_seen": 476184576 }, { "epoch": 0.14, "learning_rate": 0.0004321664994984955, "loss": 2.7657, "theoretical_loss": 3.9355181682504483, "tokens_seen": 476250112 }, { "epoch": 0.14, "learning_rate": 0.0004321564694082247, "loss": 2.8504, "theoretical_loss": 3.9354593522480803, "tokens_seen": 476315648 }, { "epoch": 0.14, "learning_rate": 0.0004321464393179539, "loss": 2.9936, "theoretical_loss": 3.93540054660315, "tokens_seen": 476381184 }, { "epoch": 0.14, "learning_rate": 0.00043213640922768306, "loss": 2.8743, "theoretical_loss": 3.935341751312408, "tokens_seen": 476446720 }, { "epoch": 0.14, "learning_rate": 0.00043212637913741224, "loss": 2.6885, "theoretical_loss": 3.9352829663726085, "tokens_seen": 476512256 }, { "epoch": 0.14, "learning_rate": 0.0004321163490471414, "loss": 2.9095, "theoretical_loss": 3.935224191780505, "tokens_seen": 476577792 }, { "epoch": 0.14, "learning_rate": 0.00043210631895687065, "loss": 2.8379, "theoretical_loss": 3.9351654275328536, "tokens_seen": 476643328 }, { "epoch": 0.14, "learning_rate": 0.0004320962888665998, "loss": 2.8026, "theoretical_loss": 3.9351066736264126, "tokens_seen": 476708864 }, { "epoch": 0.14, "objective/train/docs_used": 281398, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6347885131835938, "objective/train/theoretical_loss": 3.93504793005794, "objective/train/tokens_used": 497234400, "theoretical_loss": 3.93504793005794, "tokens_seen": 476774400 }, { "epoch": 0.14, "learning_rate": 0.000432086258776329, "loss": 2.8477, "theoretical_loss": 3.93504793005794, "tokens_seen": 476774400 }, { "epoch": 0.14, "learning_rate": 0.0004320762286860582, "loss": 2.8116, "theoretical_loss": 3.9349891968241963, "tokens_seen": 476839936 }, { "epoch": 0.14, "learning_rate": 0.0004320661985957874, "loss": 3.0569, "theoretical_loss": 3.934930473921944, "tokens_seen": 476905472 }, { "epoch": 0.14, "learning_rate": 0.00043205616850551656, "loss": 2.689, "theoretical_loss": 3.934871761347945, "tokens_seen": 476971008 }, { "epoch": 0.14, "learning_rate": 0.00043204613841524574, "loss": 2.8428, "theoretical_loss": 3.9348130590989654, "tokens_seen": 477036544 }, { "epoch": 0.14, "learning_rate": 0.0004320361083249749, "loss": 2.9822, "theoretical_loss": 3.93475436717177, "tokens_seen": 477102080 }, { "epoch": 0.14, "learning_rate": 0.00043202607823470416, "loss": 2.9795, "theoretical_loss": 3.9346956855631285, "tokens_seen": 477167616 }, { "epoch": 0.14, "learning_rate": 0.0004320160481444333, "loss": 3.0974, "theoretical_loss": 3.934637014269808, "tokens_seen": 477233152 }, { "epoch": 0.14, "learning_rate": 0.0004320060180541625, "loss": 3.0523, "theoretical_loss": 3.93457835328858, "tokens_seen": 477298688 }, { "epoch": 0.14, "learning_rate": 0.00043199598796389165, "loss": 2.8689, "theoretical_loss": 3.9345197026162158, "tokens_seen": 477364224 }, { "epoch": 0.14, "learning_rate": 0.0004319859578736209, "loss": 2.9996, "theoretical_loss": 3.9344610622494893, "tokens_seen": 477429760 }, { "epoch": 0.14, "learning_rate": 0.00043197592778335006, "loss": 2.9221, "theoretical_loss": 3.934402432185176, "tokens_seen": 477495296 }, { "epoch": 0.14, "learning_rate": 0.00043196589769307924, "loss": 2.8884, "theoretical_loss": 3.9343438124200514, "tokens_seen": 477560832 }, { "epoch": 0.14, "learning_rate": 0.0004319558676028084, "loss": 2.9777, "theoretical_loss": 3.9342852029508935, "tokens_seen": 477626368 }, { "epoch": 0.14, "learning_rate": 0.0004319458375125376, "loss": 2.7155, "theoretical_loss": 3.9342266037744817, "tokens_seen": 477691904 }, { "epoch": 0.14, "learning_rate": 0.0004319358074222668, "loss": 3.0188, "theoretical_loss": 3.934168014887597, "tokens_seen": 477757440 }, { "epoch": 0.14, "learning_rate": 0.000431925777331996, "loss": 3.0369, "theoretical_loss": 3.934109436287021, "tokens_seen": 477822976 }, { "epoch": 0.14, "learning_rate": 0.00043191574724172515, "loss": 2.8311, "theoretical_loss": 3.934050867969537, "tokens_seen": 477888512 }, { "epoch": 0.14, "learning_rate": 0.0004319057171514544, "loss": 2.9345, "theoretical_loss": 3.9339923099319316, "tokens_seen": 477954048 }, { "epoch": 0.14, "learning_rate": 0.00043189568706118357, "loss": 2.6172, "theoretical_loss": 3.93393376217099, "tokens_seen": 478019584 }, { "epoch": 0.14, "learning_rate": 0.00043188565697091275, "loss": 2.9966, "theoretical_loss": 3.9338752246835, "tokens_seen": 478085120 }, { "epoch": 0.14, "learning_rate": 0.00043187562688064193, "loss": 2.6355, "theoretical_loss": 3.933816697466252, "tokens_seen": 478150656 }, { "epoch": 0.14, "learning_rate": 0.0004318655967903711, "loss": 3.0528, "theoretical_loss": 3.933758180516036, "tokens_seen": 478216192 }, { "epoch": 0.14, "learning_rate": 0.0004318555667001003, "loss": 2.8607, "theoretical_loss": 3.9336996738296444, "tokens_seen": 478281728 }, { "epoch": 0.14, "learning_rate": 0.0004318455366098295, "loss": 3.0465, "theoretical_loss": 3.9336411774038713, "tokens_seen": 478347264 }, { "epoch": 0.14, "objective/train/docs_used": 281924, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0119645595550537, "objective/train/theoretical_loss": 3.9335826912355114, "objective/train/tokens_used": 498872800, "theoretical_loss": 3.9335826912355114, "tokens_seen": 478412800 }, { "epoch": 0.14, "learning_rate": 0.00043183550651955865, "loss": 2.8013, "theoretical_loss": 3.9335826912355114, "tokens_seen": 478412800 }, { "epoch": 0.14, "learning_rate": 0.0004318254764292879, "loss": 3.0172, "theoretical_loss": 3.9335242153213614, "tokens_seen": 478478336 }, { "epoch": 0.15, "learning_rate": 0.000431815446339017, "loss": 2.9958, "theoretical_loss": 3.9334657496582195, "tokens_seen": 478543872 }, { "epoch": 0.15, "learning_rate": 0.00043180541624874625, "loss": 2.9764, "theoretical_loss": 3.933407294242885, "tokens_seen": 478609408 }, { "epoch": 0.15, "learning_rate": 0.00043179538615847543, "loss": 2.9699, "theoretical_loss": 3.9333488490721598, "tokens_seen": 478674944 }, { "epoch": 0.15, "learning_rate": 0.0004317853560682046, "loss": 2.7685, "theoretical_loss": 3.933290414142845, "tokens_seen": 478740480 }, { "epoch": 0.15, "learning_rate": 0.0004317753259779338, "loss": 2.7827, "theoretical_loss": 3.933231989451744, "tokens_seen": 478806016 }, { "epoch": 0.15, "learning_rate": 0.00043176529588766303, "loss": 2.8662, "theoretical_loss": 3.9331735749956627, "tokens_seen": 478871552 }, { "epoch": 0.15, "learning_rate": 0.00043175526579739215, "loss": 2.8525, "theoretical_loss": 3.933115170771409, "tokens_seen": 478937088 }, { "epoch": 0.15, "learning_rate": 0.0004317452357071214, "loss": 2.8977, "theoretical_loss": 3.9330567767757887, "tokens_seen": 479002624 }, { "epoch": 0.15, "learning_rate": 0.0004317352056168505, "loss": 2.9396, "theoretical_loss": 3.932998393005613, "tokens_seen": 479068160 }, { "epoch": 0.15, "learning_rate": 0.00043172517552657975, "loss": 2.9447, "theoretical_loss": 3.932940019457692, "tokens_seen": 479133696 }, { "epoch": 0.15, "learning_rate": 0.00043171514543630893, "loss": 2.8563, "theoretical_loss": 3.9328816561288384, "tokens_seen": 479199232 }, { "epoch": 0.15, "learning_rate": 0.0004317051153460381, "loss": 2.8169, "theoretical_loss": 3.932823303015866, "tokens_seen": 479264768 }, { "epoch": 0.15, "learning_rate": 0.0004316950852557673, "loss": 2.7705, "theoretical_loss": 3.93276496011559, "tokens_seen": 479330304 }, { "epoch": 0.15, "learning_rate": 0.0004316850551654965, "loss": 3.1308, "theoretical_loss": 3.9327066274248272, "tokens_seen": 479395840 }, { "epoch": 0.15, "learning_rate": 0.0004316750250752257, "loss": 3.0848, "theoretical_loss": 3.932648304940395, "tokens_seen": 479461376 }, { "epoch": 0.15, "learning_rate": 0.0004316649949849549, "loss": 3.0146, "theoretical_loss": 3.932589992659114, "tokens_seen": 479526912 }, { "epoch": 0.15, "learning_rate": 0.0004316549648946841, "loss": 2.9719, "theoretical_loss": 3.9325316905778047, "tokens_seen": 479592448 }, { "epoch": 0.15, "learning_rate": 0.00043164493480441326, "loss": 2.9354, "theoretical_loss": 3.932473398693289, "tokens_seen": 479657984 }, { "epoch": 0.15, "learning_rate": 0.00043163490471414244, "loss": 3.2547, "theoretical_loss": 3.932415117002391, "tokens_seen": 479723520 }, { "epoch": 0.15, "learning_rate": 0.0004316248746238716, "loss": 3.0618, "theoretical_loss": 3.9323568455019364, "tokens_seen": 479789056 }, { "epoch": 0.15, "learning_rate": 0.00043161484453360085, "loss": 2.9419, "theoretical_loss": 3.932298584188751, "tokens_seen": 479854592 }, { "epoch": 0.15, "learning_rate": 0.00043160481444333, "loss": 2.9944, "theoretical_loss": 3.9322403330596636, "tokens_seen": 479920128 }, { "epoch": 0.15, "learning_rate": 0.0004315947843530592, "loss": 2.7101, "theoretical_loss": 3.9321820921115034, "tokens_seen": 479985664 }, { "epoch": 0.15, "objective/train/docs_used": 283309, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8511404991149902, "objective/train/theoretical_loss": 3.932123861341101, "objective/train/tokens_used": 500511200, "theoretical_loss": 3.932123861341101, "tokens_seen": 480051200 }, { "epoch": 0.15, "learning_rate": 0.0004315847542627884, "loss": 2.7065, "theoretical_loss": 3.932123861341101, "tokens_seen": 480051200 }, { "epoch": 0.15, "learning_rate": 0.0004315747241725176, "loss": 2.8497, "theoretical_loss": 3.932065640745289, "tokens_seen": 480116736 }, { "epoch": 0.15, "learning_rate": 0.00043156469408224676, "loss": 2.7446, "theoretical_loss": 3.932007430320901, "tokens_seen": 480182272 }, { "epoch": 0.15, "learning_rate": 0.00043155466399197594, "loss": 2.8389, "theoretical_loss": 3.9319492300647725, "tokens_seen": 480247808 }, { "epoch": 0.15, "learning_rate": 0.0004315446339017051, "loss": 2.9232, "theoretical_loss": 3.93189103997374, "tokens_seen": 480313344 }, { "epoch": 0.15, "learning_rate": 0.00043153460381143436, "loss": 2.9103, "theoretical_loss": 3.9318328600446413, "tokens_seen": 480378880 }, { "epoch": 0.15, "learning_rate": 0.0004315245737211635, "loss": 2.9765, "theoretical_loss": 3.931774690274316, "tokens_seen": 480444416 }, { "epoch": 0.15, "learning_rate": 0.0004315145436308927, "loss": 2.9858, "theoretical_loss": 3.931716530659604, "tokens_seen": 480509952 }, { "epoch": 0.15, "learning_rate": 0.00043150451354062185, "loss": 2.9393, "theoretical_loss": 3.931658381197349, "tokens_seen": 480575488 }, { "epoch": 0.15, "learning_rate": 0.0004314944834503511, "loss": 2.8405, "theoretical_loss": 3.9316002418843934, "tokens_seen": 480641024 }, { "epoch": 0.15, "learning_rate": 0.00043148445336008026, "loss": 2.9599, "theoretical_loss": 3.931542112717583, "tokens_seen": 480706560 }, { "epoch": 0.15, "learning_rate": 0.00043147442326980944, "loss": 2.7486, "theoretical_loss": 3.931483993693764, "tokens_seen": 480772096 }, { "epoch": 0.15, "learning_rate": 0.0004314643931795386, "loss": 2.998, "theoretical_loss": 3.9314258848097845, "tokens_seen": 480837632 }, { "epoch": 0.15, "learning_rate": 0.0004314543630892678, "loss": 2.8837, "theoretical_loss": 3.9313677860624936, "tokens_seen": 480903168 }, { "epoch": 0.15, "learning_rate": 0.000431444332998997, "loss": 2.9617, "theoretical_loss": 3.931309697448742, "tokens_seen": 480968704 }, { "epoch": 0.15, "learning_rate": 0.0004314343029087262, "loss": 3.1359, "theoretical_loss": 3.9312516189653817, "tokens_seen": 481034240 }, { "epoch": 0.15, "learning_rate": 0.00043142427281845535, "loss": 3.0123, "theoretical_loss": 3.931193550609266, "tokens_seen": 481099776 }, { "epoch": 0.15, "learning_rate": 0.0004314142427281846, "loss": 2.8058, "theoretical_loss": 3.9311354923772504, "tokens_seen": 481165312 }, { "epoch": 0.15, "learning_rate": 0.00043140421263791377, "loss": 2.9216, "theoretical_loss": 3.9310774442661907, "tokens_seen": 481230848 }, { "epoch": 0.15, "learning_rate": 0.00043139418254764295, "loss": 2.9538, "theoretical_loss": 3.931019406272945, "tokens_seen": 481296384 }, { "epoch": 0.15, "learning_rate": 0.00043138415245737213, "loss": 2.7984, "theoretical_loss": 3.930961378394372, "tokens_seen": 481361920 }, { "epoch": 0.15, "learning_rate": 0.0004313741223671013, "loss": 2.9692, "theoretical_loss": 3.9309033606273327, "tokens_seen": 481427456 }, { "epoch": 0.15, "learning_rate": 0.0004313640922768305, "loss": 2.8344, "theoretical_loss": 3.9308453529686886, "tokens_seen": 481492992 }, { "epoch": 0.15, "learning_rate": 0.0004313540621865597, "loss": 2.9699, "theoretical_loss": 3.930787355415304, "tokens_seen": 481558528 }, { "epoch": 0.15, "learning_rate": 0.00043134403209628885, "loss": 2.9992, "theoretical_loss": 3.930729367964042, "tokens_seen": 481624064 }, { "epoch": 0.15, "objective/train/docs_used": 283520, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7898130416870117, "objective/train/theoretical_loss": 3.93067139061177, "objective/train/tokens_used": 502149600, "theoretical_loss": 3.93067139061177, "tokens_seen": 481689600 }, { "epoch": 0.15, "learning_rate": 0.0004313340020060181, "loss": 2.9453, "theoretical_loss": 3.93067139061177, "tokens_seen": 481689600 }, { "epoch": 0.15, "learning_rate": 0.0004313239719157472, "loss": 2.9551, "theoretical_loss": 3.930613423355355, "tokens_seen": 481755136 }, { "epoch": 0.15, "learning_rate": 0.00043131394182547645, "loss": 3.2456, "theoretical_loss": 3.9305554661916657, "tokens_seen": 481820672 }, { "epoch": 0.15, "learning_rate": 0.00043130391173520563, "loss": 2.7297, "theoretical_loss": 3.9304975191175733, "tokens_seen": 481886208 }, { "epoch": 0.15, "learning_rate": 0.0004312938816449348, "loss": 3.0574, "theoretical_loss": 3.9304395821299485, "tokens_seen": 481951744 }, { "epoch": 0.15, "learning_rate": 0.000431283851554664, "loss": 2.6541, "theoretical_loss": 3.9303816552256654, "tokens_seen": 482017280 }, { "epoch": 0.15, "learning_rate": 0.00043127382146439323, "loss": 3.0918, "theoretical_loss": 3.930323738401597, "tokens_seen": 482082816 }, { "epoch": 0.15, "learning_rate": 0.00043126379137412236, "loss": 3.1726, "theoretical_loss": 3.930265831654621, "tokens_seen": 482148352 }, { "epoch": 0.15, "learning_rate": 0.0004312537612838516, "loss": 2.7858, "theoretical_loss": 3.930207934981614, "tokens_seen": 482213888 }, { "epoch": 0.15, "learning_rate": 0.0004312437311935807, "loss": 2.9732, "theoretical_loss": 3.930150048379454, "tokens_seen": 482279424 }, { "epoch": 0.15, "learning_rate": 0.00043123370110330995, "loss": 2.9883, "theoretical_loss": 3.930092171845022, "tokens_seen": 482344960 }, { "epoch": 0.15, "learning_rate": 0.00043123370110330995, "loss": 3.0501, "theoretical_loss": 3.930034305375199, "tokens_seen": 482410496 }, { "epoch": 0.15, "learning_rate": 0.00043123370110330995, "loss": 2.9022, "theoretical_loss": 3.929976448966868, "tokens_seen": 482476032 }, { "epoch": 0.15, "learning_rate": 0.00043122367101303913, "loss": 2.8998, "theoretical_loss": 3.929918602616913, "tokens_seen": 482541568 }, { "epoch": 0.15, "learning_rate": 0.0004312136409227683, "loss": 2.8981, "theoretical_loss": 3.92986076632222, "tokens_seen": 482607104 }, { "epoch": 0.15, "learning_rate": 0.0004312036108324975, "loss": 3.1734, "theoretical_loss": 3.929802940079676, "tokens_seen": 482672640 }, { "epoch": 0.15, "learning_rate": 0.0004311935807422267, "loss": 2.7899, "theoretical_loss": 3.9297451238861685, "tokens_seen": 482738176 }, { "epoch": 0.15, "learning_rate": 0.00043118355065195586, "loss": 3.0353, "theoretical_loss": 3.9296873177385887, "tokens_seen": 482803712 }, { "epoch": 0.15, "learning_rate": 0.0004311735205616851, "loss": 2.9585, "theoretical_loss": 3.929629521633827, "tokens_seen": 482869248 }, { "epoch": 0.15, "learning_rate": 0.0004311634904714142, "loss": 3.0221, "theoretical_loss": 3.929571735568776, "tokens_seen": 482934784 }, { "epoch": 0.15, "learning_rate": 0.00043115346038114346, "loss": 3.0581, "theoretical_loss": 3.92951395954033, "tokens_seen": 483000320 }, { "epoch": 0.15, "learning_rate": 0.0004311434302908726, "loss": 2.9762, "theoretical_loss": 3.9294561935453833, "tokens_seen": 483065856 }, { "epoch": 0.15, "learning_rate": 0.0004311334002006018, "loss": 3.0789, "theoretical_loss": 3.9293984375808337, "tokens_seen": 483131392 }, { "epoch": 0.15, "learning_rate": 0.000431123370110331, "loss": 2.8882, "theoretical_loss": 3.929340691643579, "tokens_seen": 483196928 }, { "epoch": 0.15, "learning_rate": 0.0004311133400200602, "loss": 2.7947, "theoretical_loss": 3.929282955730519, "tokens_seen": 483262464 }, { "epoch": 0.15, "objective/train/docs_used": 284734, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4735991954803467, "objective/train/theoretical_loss": 3.9292252298385533, "objective/train/tokens_used": 503788000, "theoretical_loss": 3.9292252298385533, "tokens_seen": 483328000 }, { "epoch": 0.15, "learning_rate": 0.00043110330992978936, "loss": 3.1294, "theoretical_loss": 3.9292252298385533, "tokens_seen": 483328000 }, { "epoch": 0.15, "learning_rate": 0.0004310932798395186, "loss": 2.921, "theoretical_loss": 3.929167513964585, "tokens_seen": 483393536 }, { "epoch": 0.15, "learning_rate": 0.0004310832497492477, "loss": 2.8072, "theoretical_loss": 3.929109808105518, "tokens_seen": 483459072 }, { "epoch": 0.15, "learning_rate": 0.00043107321965897696, "loss": 3.0061, "theoretical_loss": 3.9290521122582573, "tokens_seen": 483524608 }, { "epoch": 0.15, "learning_rate": 0.0004310631895687061, "loss": 3.1074, "theoretical_loss": 3.928994426419708, "tokens_seen": 483590144 }, { "epoch": 0.15, "learning_rate": 0.0004310531594784353, "loss": 2.9239, "theoretical_loss": 3.928936750586779, "tokens_seen": 483655680 }, { "epoch": 0.15, "learning_rate": 0.0004310431293881645, "loss": 3.066, "theoretical_loss": 3.9288790847563795, "tokens_seen": 483721216 }, { "epoch": 0.15, "learning_rate": 0.0004310330992978937, "loss": 2.9879, "theoretical_loss": 3.9288214289254193, "tokens_seen": 483786752 }, { "epoch": 0.15, "learning_rate": 0.00043102306920762286, "loss": 3.0037, "theoretical_loss": 3.9287637830908104, "tokens_seen": 483852288 }, { "epoch": 0.15, "learning_rate": 0.00043101303911735205, "loss": 2.8875, "theoretical_loss": 3.928706147249466, "tokens_seen": 483917824 }, { "epoch": 0.15, "learning_rate": 0.0004310030090270812, "loss": 2.911, "theoretical_loss": 3.9286485213983013, "tokens_seen": 483983360 }, { "epoch": 0.15, "learning_rate": 0.00043099297893681046, "loss": 2.8981, "theoretical_loss": 3.9285909055342314, "tokens_seen": 484048896 }, { "epoch": 0.15, "learning_rate": 0.0004309829488465396, "loss": 3.1723, "theoretical_loss": 3.928533299654174, "tokens_seen": 484114432 }, { "epoch": 0.15, "learning_rate": 0.0004309729187562688, "loss": 3.0433, "theoretical_loss": 3.928475703755048, "tokens_seen": 484179968 }, { "epoch": 0.15, "learning_rate": 0.00043096288866599795, "loss": 2.9147, "theoretical_loss": 3.9284181178337736, "tokens_seen": 484245504 }, { "epoch": 0.15, "learning_rate": 0.0004309528585757272, "loss": 2.8724, "theoretical_loss": 3.9283605418872716, "tokens_seen": 484311040 }, { "epoch": 0.15, "learning_rate": 0.00043094282848545637, "loss": 3.0486, "theoretical_loss": 3.9283029759124646, "tokens_seen": 484376576 }, { "epoch": 0.15, "learning_rate": 0.00043093279839518555, "loss": 3.2149, "theoretical_loss": 3.9282454199062777, "tokens_seen": 484442112 }, { "epoch": 0.15, "learning_rate": 0.0004309227683049148, "loss": 2.8778, "theoretical_loss": 3.928187873865636, "tokens_seen": 484507648 }, { "epoch": 0.15, "learning_rate": 0.00043091273821464397, "loss": 3.0867, "theoretical_loss": 3.928130337787467, "tokens_seen": 484573184 }, { "epoch": 0.15, "learning_rate": 0.00043090270812437315, "loss": 3.0863, "theoretical_loss": 3.9280728116686974, "tokens_seen": 484638720 }, { "epoch": 0.15, "learning_rate": 0.00043089267803410233, "loss": 3.154, "theoretical_loss": 3.9280152955062584, "tokens_seen": 484704256 }, { "epoch": 0.15, "learning_rate": 0.0004308826479438315, "loss": 2.9944, "theoretical_loss": 3.92795778929708, "tokens_seen": 484769792 }, { "epoch": 0.15, "learning_rate": 0.0004308726178535607, "loss": 2.9434, "theoretical_loss": 3.9279002930380953, "tokens_seen": 484835328 }, { "epoch": 0.15, "learning_rate": 0.0004308625877632899, "loss": 3.1584, "theoretical_loss": 3.927842806726237, "tokens_seen": 484900864 }, { "epoch": 0.15, "objective/train/docs_used": 285372, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4254167079925537, "objective/train/theoretical_loss": 3.927785330358441, "objective/train/tokens_used": 505426400, "theoretical_loss": 3.927785330358441, "tokens_seen": 484966400 }, { "epoch": 0.15, "learning_rate": 0.00043085255767301905, "loss": 2.8477, "theoretical_loss": 3.927785330358441, "tokens_seen": 484966400 }, { "epoch": 0.15, "learning_rate": 0.0004308425275827483, "loss": 2.9169, "theoretical_loss": 3.9277278639316435, "tokens_seen": 485031936 }, { "epoch": 0.15, "learning_rate": 0.0004308324974924774, "loss": 2.9502, "theoretical_loss": 3.9276704074427826, "tokens_seen": 485097472 }, { "epoch": 0.15, "learning_rate": 0.00043082246740220665, "loss": 2.8843, "theoretical_loss": 3.927612960888797, "tokens_seen": 485163008 }, { "epoch": 0.15, "learning_rate": 0.00043081243731193583, "loss": 2.7812, "theoretical_loss": 3.9275555242666265, "tokens_seen": 485228544 }, { "epoch": 0.15, "learning_rate": 0.000430802407221665, "loss": 3.0061, "theoretical_loss": 3.9274980975732143, "tokens_seen": 485294080 }, { "epoch": 0.15, "learning_rate": 0.0004307923771313942, "loss": 2.9978, "theoretical_loss": 3.927440680805503, "tokens_seen": 485359616 }, { "epoch": 0.15, "learning_rate": 0.00043078234704112343, "loss": 2.802, "theoretical_loss": 3.9273832739604373, "tokens_seen": 485425152 }, { "epoch": 0.15, "learning_rate": 0.00043077231695085256, "loss": 2.8012, "theoretical_loss": 3.9273258770349626, "tokens_seen": 485490688 }, { "epoch": 0.15, "learning_rate": 0.0004307622868605818, "loss": 2.9933, "theoretical_loss": 3.9272684900260266, "tokens_seen": 485556224 }, { "epoch": 0.15, "learning_rate": 0.0004307522567703109, "loss": 3.0707, "theoretical_loss": 3.9272111129305785, "tokens_seen": 485621760 }, { "epoch": 0.15, "learning_rate": 0.00043074222668004015, "loss": 3.0468, "theoretical_loss": 3.927153745745567, "tokens_seen": 485687296 }, { "epoch": 0.15, "learning_rate": 0.00043073219658976933, "loss": 3.0784, "theoretical_loss": 3.9270963884679437, "tokens_seen": 485752832 }, { "epoch": 0.15, "learning_rate": 0.0004307221664994985, "loss": 2.8892, "theoretical_loss": 3.9270390410946625, "tokens_seen": 485818368 }, { "epoch": 0.15, "learning_rate": 0.0004307121364092277, "loss": 2.9743, "theoretical_loss": 3.9269817036226757, "tokens_seen": 485883904 }, { "epoch": 0.15, "learning_rate": 0.0004307021063189569, "loss": 2.8246, "theoretical_loss": 3.92692437604894, "tokens_seen": 485949440 }, { "epoch": 0.15, "learning_rate": 0.00043069207622868606, "loss": 2.8975, "theoretical_loss": 3.9268670583704113, "tokens_seen": 486014976 }, { "epoch": 0.15, "learning_rate": 0.0004306820461384153, "loss": 3.146, "theoretical_loss": 3.926809750584048, "tokens_seen": 486080512 }, { "epoch": 0.15, "learning_rate": 0.0004306720160481444, "loss": 2.9466, "theoretical_loss": 3.9267524526868085, "tokens_seen": 486146048 }, { "epoch": 0.15, "learning_rate": 0.00043066198595787366, "loss": 3.0278, "theoretical_loss": 3.9266951646756554, "tokens_seen": 486211584 }, { "epoch": 0.15, "learning_rate": 0.0004306519558676028, "loss": 2.8523, "theoretical_loss": 3.9266378865475495, "tokens_seen": 486277120 }, { "epoch": 0.15, "learning_rate": 0.000430641925777332, "loss": 2.9906, "theoretical_loss": 3.9265806182994543, "tokens_seen": 486342656 }, { "epoch": 0.15, "learning_rate": 0.0004306318956870612, "loss": 2.8391, "theoretical_loss": 3.926523359928335, "tokens_seen": 486408192 }, { "epoch": 0.15, "learning_rate": 0.0004306218655967904, "loss": 2.8194, "theoretical_loss": 3.9264661114311576, "tokens_seen": 486473728 }, { "epoch": 0.15, "learning_rate": 0.00043061183550651956, "loss": 3.0383, "theoretical_loss": 3.926408872804889, "tokens_seen": 486539264 }, { "epoch": 0.15, "objective/train/docs_used": 285999, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2507073879241943, "objective/train/theoretical_loss": 3.9263516440464987, "objective/train/tokens_used": 507064800, "theoretical_loss": 3.9263516440464987, "tokens_seen": 486604800 }, { "epoch": 0.15, "learning_rate": 0.0004306018054162488, "loss": 2.9126, "theoretical_loss": 3.9263516440464987, "tokens_seen": 486604800 }, { "epoch": 0.15, "learning_rate": 0.0004305917753259779, "loss": 2.7026, "theoretical_loss": 3.926294425152957, "tokens_seen": 486670336 }, { "epoch": 0.15, "learning_rate": 0.00043058174523570716, "loss": 2.9969, "theoretical_loss": 3.926237216121234, "tokens_seen": 486735872 }, { "epoch": 0.15, "learning_rate": 0.0004305717151454363, "loss": 3.0097, "theoretical_loss": 3.9261800169483037, "tokens_seen": 486801408 }, { "epoch": 0.15, "learning_rate": 0.0004305616850551655, "loss": 2.9747, "theoretical_loss": 3.9261228276311404, "tokens_seen": 486866944 }, { "epoch": 0.15, "learning_rate": 0.0004305516549648947, "loss": 2.9391, "theoretical_loss": 3.926065648166719, "tokens_seen": 486932480 }, { "epoch": 0.15, "learning_rate": 0.0004305416248746239, "loss": 2.761, "theoretical_loss": 3.926008478552016, "tokens_seen": 486998016 }, { "epoch": 0.15, "learning_rate": 0.00043053159478435306, "loss": 2.8618, "theoretical_loss": 3.9259513187840103, "tokens_seen": 487063552 }, { "epoch": 0.15, "learning_rate": 0.00043052156469408225, "loss": 2.946, "theoretical_loss": 3.9258941688596805, "tokens_seen": 487129088 }, { "epoch": 0.15, "learning_rate": 0.0004305115346038114, "loss": 3.0027, "theoretical_loss": 3.9258370287760087, "tokens_seen": 487194624 }, { "epoch": 0.15, "learning_rate": 0.00043050150451354066, "loss": 3.0505, "theoretical_loss": 3.9257798985299757, "tokens_seen": 487260160 }, { "epoch": 0.15, "learning_rate": 0.0004304914744232698, "loss": 3.0067, "theoretical_loss": 3.9257227781185655, "tokens_seen": 487325696 }, { "epoch": 0.15, "learning_rate": 0.000430481444332999, "loss": 2.9225, "theoretical_loss": 3.9256656675387633, "tokens_seen": 487391232 }, { "epoch": 0.15, "learning_rate": 0.00043047141424272815, "loss": 2.8414, "theoretical_loss": 3.925608566787555, "tokens_seen": 487456768 }, { "epoch": 0.15, "learning_rate": 0.0004304613841524574, "loss": 3.0204, "theoretical_loss": 3.925551475861928, "tokens_seen": 487522304 }, { "epoch": 0.15, "learning_rate": 0.00043045135406218657, "loss": 2.9023, "theoretical_loss": 3.9254943947588705, "tokens_seen": 487587840 }, { "epoch": 0.15, "learning_rate": 0.00043044132397191575, "loss": 2.8248, "theoretical_loss": 3.9254373234753737, "tokens_seen": 487653376 }, { "epoch": 0.15, "learning_rate": 0.00043043129388164493, "loss": 3.0453, "theoretical_loss": 3.925380262008428, "tokens_seen": 487718912 }, { "epoch": 0.15, "learning_rate": 0.00043042126379137417, "loss": 3.1579, "theoretical_loss": 3.9253232103550273, "tokens_seen": 487784448 }, { "epoch": 0.15, "learning_rate": 0.0004304112337011033, "loss": 2.7528, "theoretical_loss": 3.925266168512165, "tokens_seen": 487849984 }, { "epoch": 0.15, "learning_rate": 0.00043040120361083253, "loss": 2.6098, "theoretical_loss": 3.925209136476836, "tokens_seen": 487915520 }, { "epoch": 0.15, "learning_rate": 0.00043039117352056165, "loss": 2.9598, "theoretical_loss": 3.9251521142460377, "tokens_seen": 487981056 }, { "epoch": 0.15, "learning_rate": 0.0004303811434302909, "loss": 3.0533, "theoretical_loss": 3.925095101816768, "tokens_seen": 488046592 }, { "epoch": 0.15, "learning_rate": 0.00043037111334002007, "loss": 2.87, "theoretical_loss": 3.925038099186027, "tokens_seen": 488112128 }, { "epoch": 0.15, "learning_rate": 0.00043036108324974925, "loss": 2.9433, "theoretical_loss": 3.924981106350815, "tokens_seen": 488177664 }, { "epoch": 0.15, "objective/train/docs_used": 287428, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0103299617767334, "objective/train/theoretical_loss": 3.9249241233081333, "objective/train/tokens_used": 508703200, "theoretical_loss": 3.9249241233081333, "tokens_seen": 488243200 }, { "epoch": 0.15, "learning_rate": 0.00043035105315947843, "loss": 2.7969, "theoretical_loss": 3.9249241233081333, "tokens_seen": 488243200 }, { "epoch": 0.15, "learning_rate": 0.0004303410230692076, "loss": 3.0867, "theoretical_loss": 3.924867150054985, "tokens_seen": 488308736 }, { "epoch": 0.15, "learning_rate": 0.0004303309929789368, "loss": 2.7417, "theoretical_loss": 3.9248101865883767, "tokens_seen": 488374272 }, { "epoch": 0.15, "learning_rate": 0.00043032096288866603, "loss": 2.773, "theoretical_loss": 3.9247532329053127, "tokens_seen": 488439808 }, { "epoch": 0.15, "learning_rate": 0.00043031093279839516, "loss": 2.9414, "theoretical_loss": 3.924696289002801, "tokens_seen": 488505344 }, { "epoch": 0.15, "learning_rate": 0.0004303009027081244, "loss": 2.9699, "theoretical_loss": 3.9246393548778498, "tokens_seen": 488570880 }, { "epoch": 0.15, "learning_rate": 0.0004302908726178535, "loss": 3.0303, "theoretical_loss": 3.9245824305274697, "tokens_seen": 488636416 }, { "epoch": 0.15, "learning_rate": 0.00043028084252758276, "loss": 3.1834, "theoretical_loss": 3.9245255159486714, "tokens_seen": 488701952 }, { "epoch": 0.15, "learning_rate": 0.00043027081243731194, "loss": 2.9536, "theoretical_loss": 3.924468611138467, "tokens_seen": 488767488 }, { "epoch": 0.15, "learning_rate": 0.0004302607823470411, "loss": 2.851, "theoretical_loss": 3.9244117160938714, "tokens_seen": 488833024 }, { "epoch": 0.15, "learning_rate": 0.0004302507522567703, "loss": 2.9988, "theoretical_loss": 3.9243548308118994, "tokens_seen": 488898560 }, { "epoch": 0.15, "learning_rate": 0.00043024072216649953, "loss": 2.9765, "theoretical_loss": 3.9242979552895676, "tokens_seen": 488964096 }, { "epoch": 0.15, "learning_rate": 0.00043023069207622866, "loss": 2.7578, "theoretical_loss": 3.9242410895238935, "tokens_seen": 489029632 }, { "epoch": 0.15, "learning_rate": 0.0004302206619859579, "loss": 2.8285, "theoretical_loss": 3.9241842335118964, "tokens_seen": 489095168 }, { "epoch": 0.15, "learning_rate": 0.000430210631895687, "loss": 2.7398, "theoretical_loss": 3.9241273872505973, "tokens_seen": 489160704 }, { "epoch": 0.15, "learning_rate": 0.00043020060180541626, "loss": 2.9964, "theoretical_loss": 3.924070550737017, "tokens_seen": 489226240 }, { "epoch": 0.15, "learning_rate": 0.00043019057171514544, "loss": 2.8711, "theoretical_loss": 3.9240137239681787, "tokens_seen": 489291776 }, { "epoch": 0.15, "learning_rate": 0.0004301805416248746, "loss": 3.0602, "theoretical_loss": 3.9239569069411067, "tokens_seen": 489357312 }, { "epoch": 0.15, "learning_rate": 0.00043017051153460386, "loss": 2.8675, "theoretical_loss": 3.923900099652828, "tokens_seen": 489422848 }, { "epoch": 0.15, "learning_rate": 0.000430160481444333, "loss": 2.9146, "theoretical_loss": 3.923843302100368, "tokens_seen": 489488384 }, { "epoch": 0.15, "learning_rate": 0.0004301504513540622, "loss": 3.0138, "theoretical_loss": 3.9237865142807555, "tokens_seen": 489553920 }, { "epoch": 0.15, "learning_rate": 0.0004301404212637914, "loss": 2.8976, "theoretical_loss": 3.92372973619102, "tokens_seen": 489619456 }, { "epoch": 0.15, "learning_rate": 0.0004301303911735206, "loss": 2.9251, "theoretical_loss": 3.9236729678281925, "tokens_seen": 489684992 }, { "epoch": 0.15, "learning_rate": 0.00043012036108324976, "loss": 2.8385, "theoretical_loss": 3.923616209189306, "tokens_seen": 489750528 }, { "epoch": 0.15, "learning_rate": 0.000430110330992979, "loss": 2.9459, "theoretical_loss": 3.9235594602713926, "tokens_seen": 489816064 }, { "epoch": 0.15, "objective/train/docs_used": 288116, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9460251331329346, "objective/train/theoretical_loss": 3.923502721071488, "objective/train/tokens_used": 510341600, "theoretical_loss": 3.923502721071488, "tokens_seen": 489881600 }, { "epoch": 0.15, "learning_rate": 0.0004301003009027081, "loss": 2.9832, "theoretical_loss": 3.923502721071488, "tokens_seen": 489881600 }, { "epoch": 0.15, "learning_rate": 0.00043009027081243736, "loss": 3.1877, "theoretical_loss": 3.9234459915866275, "tokens_seen": 489947136 }, { "epoch": 0.15, "learning_rate": 0.0004300802407221665, "loss": 2.753, "theoretical_loss": 3.923389271813849, "tokens_seen": 490012672 }, { "epoch": 0.15, "learning_rate": 0.0004300702106318957, "loss": 2.6838, "theoretical_loss": 3.923332561750192, "tokens_seen": 490078208 }, { "epoch": 0.15, "learning_rate": 0.0004300601805416249, "loss": 3.1324, "theoretical_loss": 3.9232758613926952, "tokens_seen": 490143744 }, { "epoch": 0.15, "learning_rate": 0.0004300501504513541, "loss": 2.8496, "theoretical_loss": 3.9232191707384, "tokens_seen": 490209280 }, { "epoch": 0.15, "learning_rate": 0.00043004012036108326, "loss": 2.9621, "theoretical_loss": 3.9231624897843496, "tokens_seen": 490274816 }, { "epoch": 0.15, "learning_rate": 0.00043003009027081245, "loss": 3.0133, "theoretical_loss": 3.9231058185275884, "tokens_seen": 490340352 }, { "epoch": 0.15, "learning_rate": 0.00043002006018054163, "loss": 2.9931, "theoretical_loss": 3.9230491569651598, "tokens_seen": 490405888 }, { "epoch": 0.15, "learning_rate": 0.00043001003009027086, "loss": 3.1411, "theoretical_loss": 3.922992505094112, "tokens_seen": 490471424 }, { "epoch": 0.15, "learning_rate": 0.00043, "loss": 2.8188, "theoretical_loss": 3.9229358629114923, "tokens_seen": 490536960 }, { "epoch": 0.15, "learning_rate": 0.0004299899699097292, "loss": 2.8421, "theoretical_loss": 3.922879230414349, "tokens_seen": 490602496 }, { "epoch": 0.15, "learning_rate": 0.00042997993981945835, "loss": 2.7538, "theoretical_loss": 3.9228226075997332, "tokens_seen": 490668032 }, { "epoch": 0.15, "learning_rate": 0.0004299699097291876, "loss": 2.8657, "theoretical_loss": 3.9227659944646964, "tokens_seen": 490733568 }, { "epoch": 0.15, "learning_rate": 0.00042995987963891677, "loss": 2.9915, "theoretical_loss": 3.9227093910062916, "tokens_seen": 490799104 }, { "epoch": 0.15, "learning_rate": 0.00042994984954864595, "loss": 3.0582, "theoretical_loss": 3.922652797221573, "tokens_seen": 490864640 }, { "epoch": 0.15, "learning_rate": 0.00042993981945837513, "loss": 2.6739, "theoretical_loss": 3.922596213107596, "tokens_seen": 490930176 }, { "epoch": 0.15, "learning_rate": 0.00042992978936810437, "loss": 3.0469, "theoretical_loss": 3.922539638661418, "tokens_seen": 490995712 }, { "epoch": 0.15, "learning_rate": 0.0004299197592778335, "loss": 3.0504, "theoretical_loss": 3.9224830738800964, "tokens_seen": 491061248 }, { "epoch": 0.15, "learning_rate": 0.00042990972918756273, "loss": 2.9923, "theoretical_loss": 3.92242651876069, "tokens_seen": 491126784 }, { "epoch": 0.15, "learning_rate": 0.00042989969909729185, "loss": 2.8921, "theoretical_loss": 3.9223699733002615, "tokens_seen": 491192320 }, { "epoch": 0.15, "learning_rate": 0.0004298896690070211, "loss": 2.944, "theoretical_loss": 3.922313437495871, "tokens_seen": 491257856 }, { "epoch": 0.15, "learning_rate": 0.00042987963891675027, "loss": 2.9777, "theoretical_loss": 3.9222569113445824, "tokens_seen": 491323392 }, { "epoch": 0.15, "learning_rate": 0.00042986960882647945, "loss": 2.9274, "theoretical_loss": 3.9222003948434603, "tokens_seen": 491388928 }, { "epoch": 0.15, "learning_rate": 0.00042985957873620863, "loss": 2.8219, "theoretical_loss": 3.92214388798957, "tokens_seen": 491454464 }, { "debugging/Self-BLEU-5": 0.43646517028931053, "debugging/distinct-1-grams": 0.8073822333747359, "debugging/distinct-2-grams": 0.976779594165329, "debugging/entropy-1-grams": 5.794474064713653, "debugging/entropy-2-grams": 6.45939179102421, "debugging/length": 536.5, "debugging/num_segments": 8, "epoch": 0.15, "objective/train/docs_used": 289197, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.5808117389678955, "objective/train/theoretical_loss": 3.92208739077998, "objective/train/tokens_used": 511980000, "theoretical_loss": 3.92208739077998, "tokens_seen": 491520000 }, { "epoch": 0.15, "learning_rate": 0.0004298495486459378, "loss": 3.1925, "theoretical_loss": 3.92208739077998, "tokens_seen": 491520000 }, { "epoch": 0.15, "learning_rate": 0.000429839518555667, "loss": 2.9464, "theoretical_loss": 3.9220309032117573, "tokens_seen": 491585536 }, { "epoch": 0.15, "learning_rate": 0.00042982948846539623, "loss": 2.8567, "theoretical_loss": 3.921974425281972, "tokens_seen": 491651072 }, { "epoch": 0.15, "learning_rate": 0.00042981945837512536, "loss": 2.9663, "theoretical_loss": 3.921917956987695, "tokens_seen": 491716608 }, { "epoch": 0.15, "learning_rate": 0.0004298094282848546, "loss": 2.9208, "theoretical_loss": 3.921861498325998, "tokens_seen": 491782144 }, { "epoch": 0.15, "learning_rate": 0.0004297993981945837, "loss": 3.063, "theoretical_loss": 3.921805049293956, "tokens_seen": 491847680 }, { "epoch": 0.15, "learning_rate": 0.00042978936810431296, "loss": 2.874, "theoretical_loss": 3.9217486098886427, "tokens_seen": 491913216 }, { "epoch": 0.15, "learning_rate": 0.00042977933801404214, "loss": 2.9789, "theoretical_loss": 3.921692180107134, "tokens_seen": 491978752 }, { "epoch": 0.15, "learning_rate": 0.0004297693079237713, "loss": 2.8321, "theoretical_loss": 3.9216357599465073, "tokens_seen": 492044288 }, { "epoch": 0.15, "learning_rate": 0.0004297592778335005, "loss": 2.8587, "theoretical_loss": 3.921579349403842, "tokens_seen": 492109824 }, { "epoch": 0.15, "learning_rate": 0.00042974924774322973, "loss": 2.9965, "theoretical_loss": 3.921522948476217, "tokens_seen": 492175360 }, { "epoch": 0.15, "learning_rate": 0.00042973921765295886, "loss": 2.9735, "theoretical_loss": 3.9214665571607146, "tokens_seen": 492240896 }, { "epoch": 0.15, "learning_rate": 0.0004297291875626881, "loss": 3.0153, "theoretical_loss": 3.921410175454416, "tokens_seen": 492306432 }, { "epoch": 0.15, "learning_rate": 0.0004297191574724172, "loss": 2.8714, "theoretical_loss": 3.921353803354406, "tokens_seen": 492371968 }, { "epoch": 0.15, "learning_rate": 0.00042970912738214646, "loss": 2.9131, "theoretical_loss": 3.9212974408577685, "tokens_seen": 492437504 }, { "epoch": 0.15, "learning_rate": 0.00042969909729187564, "loss": 2.7599, "theoretical_loss": 3.921241087961591, "tokens_seen": 492503040 }, { "epoch": 0.15, "learning_rate": 0.0004296890672016048, "loss": 2.8852, "theoretical_loss": 3.92118474466296, "tokens_seen": 492568576 }, { "epoch": 0.15, "learning_rate": 0.000429679037111334, "loss": 2.9555, "theoretical_loss": 3.9211284109589646, "tokens_seen": 492634112 }, { "epoch": 0.15, "learning_rate": 0.0004296690070210632, "loss": 2.831, "theoretical_loss": 3.9210720868466953, "tokens_seen": 492699648 }, { "epoch": 0.15, "learning_rate": 0.00042965897693079236, "loss": 3.2238, "theoretical_loss": 3.9210157723232424, "tokens_seen": 492765184 }, { "epoch": 0.15, "learning_rate": 0.0004296489468405216, "loss": 3.0512, "theoretical_loss": 3.9209594673857, "tokens_seen": 492830720 }, { "epoch": 0.15, "learning_rate": 0.0004296389167502507, "loss": 2.9424, "theoretical_loss": 3.920903172031161, "tokens_seen": 492896256 }, { "epoch": 0.15, "learning_rate": 0.00042962888665997996, "loss": 2.9471, "theoretical_loss": 3.9208468862567205, "tokens_seen": 492961792 }, { "epoch": 0.15, "learning_rate": 0.0004296188565697091, "loss": 2.8456, "theoretical_loss": 3.9207906100594756, "tokens_seen": 493027328 }, { "epoch": 0.15, "learning_rate": 0.0004296088264794383, "loss": 2.9321, "theoretical_loss": 3.920734343436523, "tokens_seen": 493092864 }, { "epoch": 0.15, "objective/train/docs_used": 289918, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.732017993927002, "objective/train/theoretical_loss": 3.920678086384963, "objective/train/tokens_used": 513618400, "theoretical_loss": 3.920678086384963, "tokens_seen": 493158400 }, { "epoch": 0.15, "learning_rate": 0.0004295987963891675, "loss": 3.0366, "theoretical_loss": 3.920678086384963, "tokens_seen": 493158400 }, { "epoch": 0.15, "learning_rate": 0.0004295887662988967, "loss": 3.0784, "theoretical_loss": 3.9206218389018943, "tokens_seen": 493223936 }, { "epoch": 0.15, "learning_rate": 0.00042957873620862587, "loss": 2.7669, "theoretical_loss": 3.9205656009844194, "tokens_seen": 493289472 }, { "epoch": 0.15, "learning_rate": 0.0004295687061183551, "loss": 3.008, "theoretical_loss": 3.9205093726296405, "tokens_seen": 493355008 }, { "epoch": 0.15, "learning_rate": 0.00042955867602808423, "loss": 2.8343, "theoretical_loss": 3.920453153834662, "tokens_seen": 493420544 }, { "epoch": 0.15, "learning_rate": 0.00042954864593781347, "loss": 2.6202, "theoretical_loss": 3.9203969445965887, "tokens_seen": 493486080 }, { "epoch": 0.15, "learning_rate": 0.0004295386158475426, "loss": 3.15, "theoretical_loss": 3.9203407449125276, "tokens_seen": 493551616 }, { "epoch": 0.15, "learning_rate": 0.00042952858575727183, "loss": 2.8284, "theoretical_loss": 3.9202845547795864, "tokens_seen": 493617152 }, { "epoch": 0.15, "learning_rate": 0.000429518555667001, "loss": 2.7332, "theoretical_loss": 3.920228374194874, "tokens_seen": 493682688 }, { "epoch": 0.15, "learning_rate": 0.0004295085255767302, "loss": 2.9725, "theoretical_loss": 3.9201722031555004, "tokens_seen": 493748224 }, { "epoch": 0.15, "learning_rate": 0.00042949849548645937, "loss": 2.9607, "theoretical_loss": 3.920116041658578, "tokens_seen": 493813760 }, { "epoch": 0.15, "learning_rate": 0.00042948846539618855, "loss": 2.7384, "theoretical_loss": 3.9200598897012187, "tokens_seen": 493879296 }, { "epoch": 0.15, "learning_rate": 0.00042947843530591773, "loss": 2.9409, "theoretical_loss": 3.9200037472805374, "tokens_seen": 493944832 }, { "epoch": 0.15, "learning_rate": 0.00042946840521564697, "loss": 2.9368, "theoretical_loss": 3.9199476143936485, "tokens_seen": 494010368 }, { "epoch": 0.15, "learning_rate": 0.0004294583751253761, "loss": 2.8874, "theoretical_loss": 3.919891491037669, "tokens_seen": 494075904 }, { "epoch": 0.15, "learning_rate": 0.00042944834503510533, "loss": 2.917, "theoretical_loss": 3.9198353772097168, "tokens_seen": 494141440 }, { "epoch": 0.15, "learning_rate": 0.0004294383149448345, "loss": 2.8774, "theoretical_loss": 3.9197792729069114, "tokens_seen": 494206976 }, { "epoch": 0.15, "learning_rate": 0.0004294282848545637, "loss": 2.9999, "theoretical_loss": 3.9197231781263726, "tokens_seen": 494272512 }, { "epoch": 0.15, "learning_rate": 0.00042941825476429293, "loss": 3.0571, "theoretical_loss": 3.919667092865222, "tokens_seen": 494338048 }, { "epoch": 0.15, "learning_rate": 0.00042940822467402206, "loss": 2.979, "theoretical_loss": 3.919611017120583, "tokens_seen": 494403584 }, { "epoch": 0.15, "learning_rate": 0.0004293981945837513, "loss": 3.0109, "theoretical_loss": 3.919554950889579, "tokens_seen": 494469120 }, { "epoch": 0.15, "learning_rate": 0.00042938816449348047, "loss": 2.9818, "theoretical_loss": 3.9194988941693354, "tokens_seen": 494534656 }, { "epoch": 0.15, "learning_rate": 0.00042937813440320965, "loss": 2.922, "theoretical_loss": 3.9194428469569793, "tokens_seen": 494600192 }, { "epoch": 0.15, "learning_rate": 0.00042936810431293883, "loss": 3.0647, "theoretical_loss": 3.9193868092496382, "tokens_seen": 494665728 }, { "epoch": 0.15, "learning_rate": 0.000429358074222668, "loss": 2.971, "theoretical_loss": 3.919330781044441, "tokens_seen": 494731264 }, { "epoch": 0.15, "objective/train/docs_used": 291263, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1252217292785645, "objective/train/theoretical_loss": 3.919274762338519, "objective/train/tokens_used": 515256800, "theoretical_loss": 3.919274762338519, "tokens_seen": 494796800 }, { "epoch": 0.15, "learning_rate": 0.0004293480441323972, "loss": 2.8186, "theoretical_loss": 3.919274762338519, "tokens_seen": 494796800 }, { "epoch": 0.15, "learning_rate": 0.00042933801404212643, "loss": 3.0019, "theoretical_loss": 3.9192187531290026, "tokens_seen": 494862336 }, { "epoch": 0.15, "learning_rate": 0.00042932798395185556, "loss": 2.8151, "theoretical_loss": 3.9191627534130253, "tokens_seen": 494927872 }, { "epoch": 0.15, "learning_rate": 0.0004293179538615848, "loss": 2.7985, "theoretical_loss": 3.9191067631877208, "tokens_seen": 494993408 }, { "epoch": 0.15, "learning_rate": 0.0004293079237713139, "loss": 3.0301, "theoretical_loss": 3.919050782450225, "tokens_seen": 495058944 }, { "epoch": 0.15, "learning_rate": 0.00042929789368104316, "loss": 2.7674, "theoretical_loss": 3.9189948111976736, "tokens_seen": 495124480 }, { "epoch": 0.15, "learning_rate": 0.00042928786359077234, "loss": 2.7445, "theoretical_loss": 3.9189388494272057, "tokens_seen": 495190016 }, { "epoch": 0.15, "learning_rate": 0.0004292778335005015, "loss": 2.9838, "theoretical_loss": 3.9188828971359584, "tokens_seen": 495255552 }, { "epoch": 0.15, "learning_rate": 0.0004292678034102307, "loss": 2.9377, "theoretical_loss": 3.918826954321074, "tokens_seen": 495321088 }, { "epoch": 0.15, "learning_rate": 0.00042925777331995993, "loss": 2.9266, "theoretical_loss": 3.918771020979693, "tokens_seen": 495386624 }, { "epoch": 0.15, "learning_rate": 0.00042924774322968906, "loss": 2.7544, "theoretical_loss": 3.9187150971089584, "tokens_seen": 495452160 }, { "epoch": 0.15, "learning_rate": 0.0004292377131394183, "loss": 3.0238, "theoretical_loss": 3.918659182706014, "tokens_seen": 495517696 }, { "epoch": 0.15, "learning_rate": 0.0004292276830491474, "loss": 2.9448, "theoretical_loss": 3.9186032777680055, "tokens_seen": 495583232 }, { "epoch": 0.15, "learning_rate": 0.00042921765295887666, "loss": 3.021, "theoretical_loss": 3.918547382292079, "tokens_seen": 495648768 }, { "epoch": 0.15, "learning_rate": 0.00042920762286860584, "loss": 2.8781, "theoretical_loss": 3.918491496275383, "tokens_seen": 495714304 }, { "epoch": 0.15, "learning_rate": 0.000429197592778335, "loss": 2.764, "theoretical_loss": 3.9184356197150647, "tokens_seen": 495779840 }, { "epoch": 0.15, "learning_rate": 0.0004291875626880642, "loss": 2.8178, "theoretical_loss": 3.918379752608276, "tokens_seen": 495845376 }, { "epoch": 0.15, "learning_rate": 0.0004291775325977934, "loss": 3.175, "theoretical_loss": 3.918323894952168, "tokens_seen": 495910912 }, { "epoch": 0.15, "learning_rate": 0.00042916750250752256, "loss": 3.0742, "theoretical_loss": 3.9182680467438926, "tokens_seen": 495976448 }, { "epoch": 0.15, "learning_rate": 0.0004291574724172518, "loss": 3.1731, "theoretical_loss": 3.9182122079806048, "tokens_seen": 496041984 }, { "epoch": 0.15, "learning_rate": 0.0004291474423269809, "loss": 2.8936, "theoretical_loss": 3.918156378659459, "tokens_seen": 496107520 }, { "epoch": 0.15, "learning_rate": 0.00042913741223671016, "loss": 2.8979, "theoretical_loss": 3.9181005587776117, "tokens_seen": 496173056 }, { "epoch": 0.15, "learning_rate": 0.0004291273821464393, "loss": 2.7485, "theoretical_loss": 3.9180447483322203, "tokens_seen": 496238592 }, { "epoch": 0.15, "learning_rate": 0.0004291173520561685, "loss": 2.7746, "theoretical_loss": 3.9179889473204446, "tokens_seen": 496304128 }, { "epoch": 0.15, "learning_rate": 0.0004291073219658977, "loss": 3.2393, "theoretical_loss": 3.9179331557394432, "tokens_seen": 496369664 }, { "epoch": 0.15, "objective/train/docs_used": 291574, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5555975437164307, "objective/train/theoretical_loss": 3.9178773735863786, "objective/train/tokens_used": 516895200, "theoretical_loss": 3.9178773735863786, "tokens_seen": 496435200 }, { "epoch": 0.15, "learning_rate": 0.0004290972918756269, "loss": 2.9531, "theoretical_loss": 3.9178773735863786, "tokens_seen": 496435200 }, { "epoch": 0.15, "learning_rate": 0.00042908726178535607, "loss": 3.0201, "theoretical_loss": 3.917821600858413, "tokens_seen": 496500736 }, { "epoch": 0.15, "learning_rate": 0.0004290772316950853, "loss": 3.0068, "theoretical_loss": 3.917765837552709, "tokens_seen": 496566272 }, { "epoch": 0.15, "learning_rate": 0.00042906720160481443, "loss": 2.9994, "theoretical_loss": 3.917710083666434, "tokens_seen": 496631808 }, { "epoch": 0.15, "learning_rate": 0.00042905717151454367, "loss": 2.8756, "theoretical_loss": 3.917654339196752, "tokens_seen": 496697344 }, { "epoch": 0.15, "learning_rate": 0.0004290471414242728, "loss": 3.1561, "theoretical_loss": 3.9175986041408315, "tokens_seen": 496762880 }, { "epoch": 0.15, "learning_rate": 0.00042903711133400203, "loss": 3.0554, "theoretical_loss": 3.9175428784958406, "tokens_seen": 496828416 }, { "epoch": 0.15, "learning_rate": 0.0004290270812437312, "loss": 2.9631, "theoretical_loss": 3.9174871622589498, "tokens_seen": 496893952 }, { "epoch": 0.15, "learning_rate": 0.0004290170511534604, "loss": 3.1769, "theoretical_loss": 3.91743145542733, "tokens_seen": 496959488 }, { "epoch": 0.15, "learning_rate": 0.00042900702106318957, "loss": 2.9962, "theoretical_loss": 3.9173757579981534, "tokens_seen": 497025024 }, { "epoch": 0.15, "learning_rate": 0.00042899699097291875, "loss": 2.9916, "theoretical_loss": 3.9173200699685937, "tokens_seen": 497090560 }, { "epoch": 0.15, "learning_rate": 0.00042898696088264793, "loss": 2.9978, "theoretical_loss": 3.9172643913358254, "tokens_seen": 497156096 }, { "epoch": 0.15, "learning_rate": 0.00042897693079237717, "loss": 2.8487, "theoretical_loss": 3.9172087220970244, "tokens_seen": 497221632 }, { "epoch": 0.15, "learning_rate": 0.0004289669007021063, "loss": 2.8799, "theoretical_loss": 3.917153062249368, "tokens_seen": 497287168 }, { "epoch": 0.15, "learning_rate": 0.00042895687061183553, "loss": 2.991, "theoretical_loss": 3.917097411790036, "tokens_seen": 497352704 }, { "epoch": 0.15, "learning_rate": 0.0004289468405215647, "loss": 2.7169, "theoretical_loss": 3.9170417707162057, "tokens_seen": 497418240 }, { "epoch": 0.15, "learning_rate": 0.0004289368104312939, "loss": 2.988, "theoretical_loss": 3.9169861390250595, "tokens_seen": 497483776 }, { "epoch": 0.15, "learning_rate": 0.0004289267803410231, "loss": 3.1274, "theoretical_loss": 3.9169305167137796, "tokens_seen": 497549312 }, { "epoch": 0.15, "learning_rate": 0.00042891675025075226, "loss": 2.8603, "theoretical_loss": 3.916874903779549, "tokens_seen": 497614848 }, { "epoch": 0.15, "learning_rate": 0.00042890672016048144, "loss": 3.3354, "theoretical_loss": 3.9168193002195513, "tokens_seen": 497680384 }, { "epoch": 0.15, "learning_rate": 0.00042889669007021067, "loss": 3.0345, "theoretical_loss": 3.9167637060309732, "tokens_seen": 497745920 }, { "epoch": 0.15, "learning_rate": 0.0004288866599799398, "loss": 2.9234, "theoretical_loss": 3.916708121211002, "tokens_seen": 497811456 }, { "epoch": 0.15, "learning_rate": 0.00042887662988966903, "loss": 2.9288, "theoretical_loss": 3.9166525457568246, "tokens_seen": 497876992 }, { "epoch": 0.15, "learning_rate": 0.00042886659979939816, "loss": 2.7802, "theoretical_loss": 3.916596979665632, "tokens_seen": 497942528 }, { "epoch": 0.15, "learning_rate": 0.0004288565697091274, "loss": 3.1781, "theoretical_loss": 3.9165414229346136, "tokens_seen": 498008064 }, { "epoch": 0.15, "objective/train/docs_used": 293024, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.047236204147339, "objective/train/theoretical_loss": 3.9164858755609613, "objective/train/tokens_used": 518533600, "theoretical_loss": 3.9164858755609613, "tokens_seen": 498073600 }, { "epoch": 0.15, "learning_rate": 0.0004288465396188566, "loss": 2.9409, "theoretical_loss": 3.9164858755609613, "tokens_seen": 498073600 }, { "epoch": 0.15, "learning_rate": 0.00042883650952858576, "loss": 3.1637, "theoretical_loss": 3.916430337541869, "tokens_seen": 498139136 }, { "epoch": 0.15, "learning_rate": 0.00042882647943831494, "loss": 3.1612, "theoretical_loss": 3.9163748088745303, "tokens_seen": 498204672 }, { "epoch": 0.15, "learning_rate": 0.0004288164493480441, "loss": 3.1534, "theoretical_loss": 3.9163192895561405, "tokens_seen": 498270208 }, { "epoch": 0.15, "learning_rate": 0.0004288064192577733, "loss": 3.0411, "theoretical_loss": 3.916263779583896, "tokens_seen": 498335744 }, { "epoch": 0.15, "learning_rate": 0.00042879638916750254, "loss": 3.1491, "theoretical_loss": 3.9162082789549952, "tokens_seen": 498401280 }, { "epoch": 0.15, "learning_rate": 0.00042878635907723166, "loss": 3.0577, "theoretical_loss": 3.916152787666638, "tokens_seen": 498466816 }, { "epoch": 0.15, "learning_rate": 0.0004287763289869609, "loss": 2.9438, "theoretical_loss": 3.916097305716023, "tokens_seen": 498532352 }, { "epoch": 0.15, "learning_rate": 0.0004287662988966901, "loss": 2.9937, "theoretical_loss": 3.9160418331003526, "tokens_seen": 498597888 }, { "epoch": 0.15, "learning_rate": 0.00042875626880641926, "loss": 2.8485, "theoretical_loss": 3.91598636981683, "tokens_seen": 498663424 }, { "epoch": 0.15, "learning_rate": 0.00042874623871614844, "loss": 2.9531, "theoretical_loss": 3.9159309158626576, "tokens_seen": 498728960 }, { "epoch": 0.15, "learning_rate": 0.0004287362086258776, "loss": 3.048, "theoretical_loss": 3.9158754712350414, "tokens_seen": 498794496 }, { "epoch": 0.15, "learning_rate": 0.0004287261785356068, "loss": 2.8666, "theoretical_loss": 3.915820035931188, "tokens_seen": 498860032 }, { "epoch": 0.15, "learning_rate": 0.00042871614844533604, "loss": 3.0475, "theoretical_loss": 3.9157646099483046, "tokens_seen": 498925568 }, { "epoch": 0.15, "learning_rate": 0.00042870611835506517, "loss": 2.6643, "theoretical_loss": 3.9157091932836, "tokens_seen": 498991104 }, { "epoch": 0.15, "learning_rate": 0.0004286960882647944, "loss": 2.9152, "theoretical_loss": 3.9156537859342837, "tokens_seen": 499056640 }, { "epoch": 0.15, "learning_rate": 0.00042868605817452353, "loss": 2.8303, "theoretical_loss": 3.9155983878975675, "tokens_seen": 499122176 }, { "epoch": 0.15, "learning_rate": 0.00042867602808425276, "loss": 3.2016, "theoretical_loss": 3.9155429991706634, "tokens_seen": 499187712 }, { "epoch": 0.15, "learning_rate": 0.000428665997993982, "loss": 3.1577, "theoretical_loss": 3.9154876197507846, "tokens_seen": 499253248 }, { "epoch": 0.15, "learning_rate": 0.0004286559679037111, "loss": 3.148, "theoretical_loss": 3.915432249635147, "tokens_seen": 499318784 }, { "epoch": 0.15, "learning_rate": 0.00042864593781344036, "loss": 2.9411, "theoretical_loss": 3.9153768888209646, "tokens_seen": 499384320 }, { "epoch": 0.15, "learning_rate": 0.0004286359077231695, "loss": 3.1934, "theoretical_loss": 3.915321537305456, "tokens_seen": 499449856 }, { "epoch": 0.15, "learning_rate": 0.0004286258776328987, "loss": 3.0267, "theoretical_loss": 3.915266195085839, "tokens_seen": 499515392 }, { "epoch": 0.15, "learning_rate": 0.0004286158475426279, "loss": 2.9095, "theoretical_loss": 3.9152108621593333, "tokens_seen": 499580928 }, { "epoch": 0.15, "learning_rate": 0.0004286058174523571, "loss": 3.3386, "theoretical_loss": 3.91515553852316, "tokens_seen": 499646464 }, { "epoch": 0.15, "objective/train/docs_used": 293514, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7597315311431885, "objective/train/theoretical_loss": 3.91510022417454, "objective/train/tokens_used": 520172000, "theoretical_loss": 3.91510022417454, "tokens_seen": 499712000 }, { "epoch": 0.15, "learning_rate": 0.00042859578736208627, "loss": 3.0573, "theoretical_loss": 3.91510022417454, "tokens_seen": 499712000 }, { "epoch": 0.15, "learning_rate": 0.0004285857572718155, "loss": 3.036, "theoretical_loss": 3.915044919110697, "tokens_seen": 499777536 }, { "epoch": 0.15, "learning_rate": 0.00042857572718154463, "loss": 2.9206, "theoretical_loss": 3.9149896233288555, "tokens_seen": 499843072 }, { "epoch": 0.15, "learning_rate": 0.00042856569709127387, "loss": 3.0611, "theoretical_loss": 3.9149343368262404, "tokens_seen": 499908608 }, { "epoch": 0.15, "learning_rate": 0.000428555667001003, "loss": 2.884, "theoretical_loss": 3.9148790596000786, "tokens_seen": 499974144 }, { "epoch": 0.15, "learning_rate": 0.00042854563691073223, "loss": 2.8042, "theoretical_loss": 3.9148237916475983, "tokens_seen": 500039680 }, { "epoch": 0.15, "learning_rate": 0.0004285356068204614, "loss": 3.0384, "theoretical_loss": 3.9147685329660282, "tokens_seen": 500105216 }, { "epoch": 0.15, "learning_rate": 0.0004285255767301906, "loss": 2.9233, "theoretical_loss": 3.914713283552599, "tokens_seen": 500170752 }, { "epoch": 0.15, "learning_rate": 0.00042851554663991977, "loss": 3.1468, "theoretical_loss": 3.9146580434045415, "tokens_seen": 500236288 }, { "epoch": 0.15, "learning_rate": 0.00042850551654964895, "loss": 3.0112, "theoretical_loss": 3.914602812519089, "tokens_seen": 500301824 }, { "epoch": 0.15, "learning_rate": 0.00042849548645937813, "loss": 2.71, "theoretical_loss": 3.9145475908934753, "tokens_seen": 500367360 }, { "epoch": 0.15, "learning_rate": 0.00042848545636910737, "loss": 3.0124, "theoretical_loss": 3.914492378524934, "tokens_seen": 500432896 }, { "epoch": 0.15, "learning_rate": 0.0004284754262788365, "loss": 2.7896, "theoretical_loss": 3.9144371754107032, "tokens_seen": 500498432 }, { "epoch": 0.15, "learning_rate": 0.00042846539618856573, "loss": 3.3801, "theoretical_loss": 3.9143819815480194, "tokens_seen": 500563968 }, { "epoch": 0.15, "learning_rate": 0.0004284553660982949, "loss": 3.0366, "theoretical_loss": 3.9143267969341218, "tokens_seen": 500629504 }, { "epoch": 0.15, "learning_rate": 0.0004284453360080241, "loss": 2.8845, "theoretical_loss": 3.9142716215662494, "tokens_seen": 500695040 }, { "epoch": 0.15, "learning_rate": 0.0004284353059177533, "loss": 2.9944, "theoretical_loss": 3.9142164554416428, "tokens_seen": 500760576 }, { "epoch": 0.15, "learning_rate": 0.00042842527582748246, "loss": 2.8842, "theoretical_loss": 3.9141612985575454, "tokens_seen": 500826112 }, { "epoch": 0.15, "learning_rate": 0.00042841524573721164, "loss": 3.0182, "theoretical_loss": 3.9141061509111994, "tokens_seen": 500891648 }, { "epoch": 0.15, "learning_rate": 0.00042840521564694087, "loss": 2.9012, "theoretical_loss": 3.91405101249985, "tokens_seen": 500957184 }, { "epoch": 0.15, "learning_rate": 0.00042839518555667, "loss": 3.0928, "theoretical_loss": 3.9139958833207427, "tokens_seen": 501022720 }, { "epoch": 0.15, "learning_rate": 0.00042838515546639923, "loss": 2.9251, "theoretical_loss": 3.9139407633711243, "tokens_seen": 501088256 }, { "epoch": 0.15, "learning_rate": 0.00042837512537612836, "loss": 3.2046, "theoretical_loss": 3.9138856526482426, "tokens_seen": 501153792 }, { "epoch": 0.15, "learning_rate": 0.0004283650952858576, "loss": 2.9961, "theoretical_loss": 3.9138305511493474, "tokens_seen": 501219328 }, { "epoch": 0.15, "learning_rate": 0.0004283550651955868, "loss": 3.1201, "theoretical_loss": 3.9137754588716884, "tokens_seen": 501284864 }, { "epoch": 0.15, "objective/train/docs_used": 294722, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9437036514282227, "objective/train/theoretical_loss": 3.9137203758125176, "objective/train/tokens_used": 521810400, "theoretical_loss": 3.9137203758125176, "tokens_seen": 501350400 }, { "epoch": 0.15, "learning_rate": 0.00042834503510531596, "loss": 2.9287, "theoretical_loss": 3.9137203758125176, "tokens_seen": 501350400 }, { "epoch": 0.15, "learning_rate": 0.00042833500501504514, "loss": 3.0112, "theoretical_loss": 3.913665301969088, "tokens_seen": 501415936 }, { "epoch": 0.15, "learning_rate": 0.0004283249749247743, "loss": 2.9346, "theoretical_loss": 3.913610237338653, "tokens_seen": 501481472 }, { "epoch": 0.15, "learning_rate": 0.0004283149448345035, "loss": 2.9928, "theoretical_loss": 3.9135551819184675, "tokens_seen": 501547008 }, { "epoch": 0.15, "learning_rate": 0.00042830491474423274, "loss": 2.9063, "theoretical_loss": 3.913500135705789, "tokens_seen": 501612544 }, { "epoch": 0.15, "learning_rate": 0.00042829488465396186, "loss": 2.83, "theoretical_loss": 3.9134450986978737, "tokens_seen": 501678080 }, { "epoch": 0.15, "learning_rate": 0.0004282848545636911, "loss": 2.7924, "theoretical_loss": 3.913390070891981, "tokens_seen": 501743616 }, { "epoch": 0.15, "learning_rate": 0.0004282748244734203, "loss": 3.0962, "theoretical_loss": 3.9133350522853703, "tokens_seen": 501809152 }, { "epoch": 0.15, "learning_rate": 0.00042826479438314946, "loss": 2.7333, "theoretical_loss": 3.9132800428753027, "tokens_seen": 501874688 }, { "epoch": 0.15, "learning_rate": 0.00042825476429287864, "loss": 2.9528, "theoretical_loss": 3.9132250426590405, "tokens_seen": 501940224 }, { "epoch": 0.15, "learning_rate": 0.0004282447342026078, "loss": 2.9171, "theoretical_loss": 3.9131700516338475, "tokens_seen": 502005760 }, { "epoch": 0.15, "learning_rate": 0.000428234704112337, "loss": 3.0443, "theoretical_loss": 3.9131150697969868, "tokens_seen": 502071296 }, { "epoch": 0.15, "learning_rate": 0.00042822467402206624, "loss": 3.1714, "theoretical_loss": 3.913060097145725, "tokens_seen": 502136832 }, { "epoch": 0.15, "learning_rate": 0.00042821464393179537, "loss": 3.0249, "theoretical_loss": 3.913005133677329, "tokens_seen": 502202368 }, { "epoch": 0.15, "learning_rate": 0.0004282046138415246, "loss": 2.9623, "theoretical_loss": 3.9129501793890666, "tokens_seen": 502267904 }, { "epoch": 0.15, "learning_rate": 0.00042819458375125373, "loss": 2.9646, "theoretical_loss": 3.9128952342782073, "tokens_seen": 502333440 }, { "epoch": 0.15, "learning_rate": 0.00042818455366098296, "loss": 2.9892, "theoretical_loss": 3.9128402983420214, "tokens_seen": 502398976 }, { "epoch": 0.15, "learning_rate": 0.00042817452357071215, "loss": 2.7224, "theoretical_loss": 3.91278537157778, "tokens_seen": 502464512 }, { "epoch": 0.15, "learning_rate": 0.0004281644934804413, "loss": 2.9322, "theoretical_loss": 3.912730453982756, "tokens_seen": 502530048 }, { "epoch": 0.15, "learning_rate": 0.0004281544633901705, "loss": 2.8625, "theoretical_loss": 3.9126755455542233, "tokens_seen": 502595584 }, { "epoch": 0.15, "learning_rate": 0.0004281444332998997, "loss": 2.7713, "theoretical_loss": 3.912620646289457, "tokens_seen": 502661120 }, { "epoch": 0.15, "learning_rate": 0.00042813440320962887, "loss": 3.0138, "theoretical_loss": 3.9125657561857334, "tokens_seen": 502726656 }, { "epoch": 0.15, "learning_rate": 0.0004281243731193581, "loss": 3.0685, "theoretical_loss": 3.9125108752403293, "tokens_seen": 502792192 }, { "epoch": 0.15, "learning_rate": 0.00042811434302908723, "loss": 2.96, "theoretical_loss": 3.9124560034505236, "tokens_seen": 502857728 }, { "epoch": 0.15, "learning_rate": 0.00042810431293881647, "loss": 2.7711, "theoretical_loss": 3.9124011408135964, "tokens_seen": 502923264 }, { "epoch": 0.15, "objective/train/docs_used": 295112, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0761241912841797, "objective/train/theoretical_loss": 3.9123462873268275, "objective/train/tokens_used": 523448800, "theoretical_loss": 3.9123462873268275, "tokens_seen": 502988800 }, { "epoch": 0.15, "learning_rate": 0.00042809428284854565, "loss": 2.8803, "theoretical_loss": 3.9123462873268275, "tokens_seen": 502988800 }, { "epoch": 0.15, "learning_rate": 0.00042808425275827483, "loss": 3.0619, "theoretical_loss": 3.9122914429874998, "tokens_seen": 503054336 }, { "epoch": 0.15, "learning_rate": 0.000428074222668004, "loss": 2.8147, "theoretical_loss": 3.912236607792896, "tokens_seen": 503119872 }, { "epoch": 0.15, "learning_rate": 0.0004280641925777332, "loss": 3.1577, "theoretical_loss": 3.9121817817403004, "tokens_seen": 503185408 }, { "epoch": 0.15, "learning_rate": 0.0004280541624874624, "loss": 3.0249, "theoretical_loss": 3.912126964826999, "tokens_seen": 503250944 }, { "epoch": 0.15, "learning_rate": 0.0004280441323971916, "loss": 2.7746, "theoretical_loss": 3.912072157050278, "tokens_seen": 503316480 }, { "epoch": 0.15, "learning_rate": 0.00042803410230692074, "loss": 3.0252, "theoretical_loss": 3.9120173584074256, "tokens_seen": 503382016 }, { "epoch": 0.15, "learning_rate": 0.00042802407221664997, "loss": 2.9913, "theoretical_loss": 3.91196256889573, "tokens_seen": 503447552 }, { "epoch": 0.15, "learning_rate": 0.0004280140421263791, "loss": 2.9239, "theoretical_loss": 3.911907788512482, "tokens_seen": 503513088 }, { "epoch": 0.15, "learning_rate": 0.00042800401203610833, "loss": 2.9242, "theoretical_loss": 3.911853017254973, "tokens_seen": 503578624 }, { "epoch": 0.15, "learning_rate": 0.0004279939819458375, "loss": 3.0287, "theoretical_loss": 3.9117982551204946, "tokens_seen": 503644160 }, { "epoch": 0.15, "learning_rate": 0.0004279839518555667, "loss": 3.0197, "theoretical_loss": 3.911743502106341, "tokens_seen": 503709696 }, { "epoch": 0.15, "learning_rate": 0.0004279739217652959, "loss": 2.8153, "theoretical_loss": 3.911688758209807, "tokens_seen": 503775232 }, { "epoch": 0.15, "learning_rate": 0.0004279638916750251, "loss": 2.93, "theoretical_loss": 3.9116340234281886, "tokens_seen": 503840768 }, { "epoch": 0.15, "learning_rate": 0.00042795386158475424, "loss": 3.1179, "theoretical_loss": 3.911579297758782, "tokens_seen": 503906304 }, { "epoch": 0.15, "learning_rate": 0.0004279438314944835, "loss": 2.7839, "theoretical_loss": 3.9115245811988864, "tokens_seen": 503971840 }, { "epoch": 0.15, "learning_rate": 0.0004279338014042126, "loss": 2.991, "theoretical_loss": 3.9114698737458005, "tokens_seen": 504037376 }, { "epoch": 0.15, "learning_rate": 0.00042792377131394184, "loss": 3.0795, "theoretical_loss": 3.911415175396825, "tokens_seen": 504102912 }, { "epoch": 0.15, "learning_rate": 0.00042791374122367107, "loss": 2.9079, "theoretical_loss": 3.911360486149262, "tokens_seen": 504168448 }, { "epoch": 0.15, "learning_rate": 0.0004279037111334002, "loss": 2.9344, "theoretical_loss": 3.9113058060004136, "tokens_seen": 504233984 }, { "epoch": 0.15, "learning_rate": 0.00042789368104312943, "loss": 2.957, "theoretical_loss": 3.9112511349475843, "tokens_seen": 504299520 }, { "epoch": 0.15, "learning_rate": 0.00042788365095285856, "loss": 2.9757, "theoretical_loss": 3.911196472988079, "tokens_seen": 504365056 }, { "epoch": 0.15, "learning_rate": 0.0004278736208625878, "loss": 3.0768, "theoretical_loss": 3.9111418201192034, "tokens_seen": 504430592 }, { "epoch": 0.15, "learning_rate": 0.000427863590772317, "loss": 2.698, "theoretical_loss": 3.911087176338266, "tokens_seen": 504496128 }, { "epoch": 0.15, "learning_rate": 0.00042785356068204616, "loss": 3.0198, "theoretical_loss": 3.911032541642575, "tokens_seen": 504561664 }, { "epoch": 0.15, "objective/train/docs_used": 296352, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.776776075363159, "objective/train/theoretical_loss": 3.910977916029439, "objective/train/tokens_used": 525087200, "theoretical_loss": 3.910977916029439, "tokens_seen": 504627200 }, { "epoch": 0.15, "learning_rate": 0.00042784353059177534, "loss": 2.8536, "theoretical_loss": 3.910977916029439, "tokens_seen": 504627200 }, { "epoch": 0.15, "learning_rate": 0.0004278335005015045, "loss": 3.1104, "theoretical_loss": 3.9109232994961705, "tokens_seen": 504692736 }, { "epoch": 0.15, "learning_rate": 0.0004278234704112337, "loss": 3.0926, "theoretical_loss": 3.91086869204008, "tokens_seen": 504758272 }, { "epoch": 0.15, "learning_rate": 0.00042781344032096294, "loss": 3.0256, "theoretical_loss": 3.910814093658482, "tokens_seen": 504823808 }, { "epoch": 0.15, "learning_rate": 0.00042780341023069206, "loss": 2.9688, "theoretical_loss": 3.9107595043486905, "tokens_seen": 504889344 }, { "epoch": 0.15, "learning_rate": 0.0004277933801404213, "loss": 3.0209, "theoretical_loss": 3.91070492410802, "tokens_seen": 504954880 }, { "epoch": 0.15, "learning_rate": 0.0004277833500501505, "loss": 2.9499, "theoretical_loss": 3.9106503529337875, "tokens_seen": 505020416 }, { "epoch": 0.15, "learning_rate": 0.00042777331995987966, "loss": 3.0091, "theoretical_loss": 3.9105957908233107, "tokens_seen": 505085952 }, { "epoch": 0.15, "learning_rate": 0.00042776328986960884, "loss": 2.6414, "theoretical_loss": 3.910541237773909, "tokens_seen": 505151488 }, { "epoch": 0.15, "learning_rate": 0.000427753259779338, "loss": 3.005, "theoretical_loss": 3.910486693782902, "tokens_seen": 505217024 }, { "epoch": 0.15, "learning_rate": 0.0004277432296890672, "loss": 2.9431, "theoretical_loss": 3.9104321588476108, "tokens_seen": 505282560 }, { "epoch": 0.15, "learning_rate": 0.00042773319959879644, "loss": 3.1832, "theoretical_loss": 3.910377632965357, "tokens_seen": 505348096 }, { "epoch": 0.15, "learning_rate": 0.00042772316950852557, "loss": 2.9311, "theoretical_loss": 3.9103231161334655, "tokens_seen": 505413632 }, { "epoch": 0.15, "learning_rate": 0.0004277131394182548, "loss": 2.8145, "theoretical_loss": 3.9102686083492593, "tokens_seen": 505479168 }, { "epoch": 0.15, "learning_rate": 0.00042770310932798393, "loss": 2.8171, "theoretical_loss": 3.910214109610065, "tokens_seen": 505544704 }, { "epoch": 0.15, "learning_rate": 0.00042769307923771316, "loss": 2.9648, "theoretical_loss": 3.910159619913209, "tokens_seen": 505610240 }, { "epoch": 0.15, "learning_rate": 0.00042768304914744235, "loss": 3.1846, "theoretical_loss": 3.9101051392560198, "tokens_seen": 505675776 }, { "epoch": 0.15, "learning_rate": 0.00042767301905717153, "loss": 2.9035, "theoretical_loss": 3.9100506676358266, "tokens_seen": 505741312 }, { "epoch": 0.15, "learning_rate": 0.0004276629889669007, "loss": 2.9588, "theoretical_loss": 3.9099962050499584, "tokens_seen": 505806848 }, { "epoch": 0.15, "learning_rate": 0.0004276529588766299, "loss": 2.923, "theoretical_loss": 3.9099417514957473, "tokens_seen": 505872384 }, { "epoch": 0.15, "learning_rate": 0.00042764292878635907, "loss": 2.8175, "theoretical_loss": 3.9098873069705267, "tokens_seen": 505937920 }, { "epoch": 0.15, "learning_rate": 0.0004276328986960883, "loss": 2.9874, "theoretical_loss": 3.9098328714716284, "tokens_seen": 506003456 }, { "epoch": 0.15, "learning_rate": 0.00042762286860581743, "loss": 3.1272, "theoretical_loss": 3.909778444996389, "tokens_seen": 506068992 }, { "epoch": 0.15, "learning_rate": 0.00042761283851554667, "loss": 3.181, "theoretical_loss": 3.9097240275421425, "tokens_seen": 506134528 }, { "epoch": 0.15, "learning_rate": 0.00042760280842527585, "loss": 2.7998, "theoretical_loss": 3.9096696191062277, "tokens_seen": 506200064 }, { "epoch": 0.15, "objective/train/docs_used": 297118, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.69783616065979, "objective/train/theoretical_loss": 3.909615219685982, "objective/train/tokens_used": 526725600, "theoretical_loss": 3.909615219685982, "tokens_seen": 506265600 }, { "epoch": 0.15, "learning_rate": 0.00042759277833500503, "loss": 2.9313, "theoretical_loss": 3.909615219685982, "tokens_seen": 506265600 }, { "epoch": 0.15, "learning_rate": 0.0004275827482447342, "loss": 2.841, "theoretical_loss": 3.9095608292787443, "tokens_seen": 506331136 }, { "epoch": 0.15, "learning_rate": 0.0004275727181544634, "loss": 3.0147, "theoretical_loss": 3.9095064478818555, "tokens_seen": 506396672 }, { "epoch": 0.15, "learning_rate": 0.0004275626880641926, "loss": 2.901, "theoretical_loss": 3.9094520754926574, "tokens_seen": 506462208 }, { "epoch": 0.15, "learning_rate": 0.0004275526579739218, "loss": 2.8451, "theoretical_loss": 3.9093977121084915, "tokens_seen": 506527744 }, { "epoch": 0.15, "learning_rate": 0.00042754262788365094, "loss": 2.9748, "theoretical_loss": 3.909343357726703, "tokens_seen": 506593280 }, { "epoch": 0.15, "learning_rate": 0.00042753259779338017, "loss": 2.9838, "theoretical_loss": 3.9092890123446367, "tokens_seen": 506658816 }, { "epoch": 0.15, "learning_rate": 0.0004275225677031093, "loss": 2.9351, "theoretical_loss": 3.9092346759596373, "tokens_seen": 506724352 }, { "epoch": 0.15, "learning_rate": 0.00042751253761283853, "loss": 2.9645, "theoretical_loss": 3.9091803485690537, "tokens_seen": 506789888 }, { "epoch": 0.15, "learning_rate": 0.0004275025075225677, "loss": 3.053, "theoretical_loss": 3.9091260301702326, "tokens_seen": 506855424 }, { "epoch": 0.15, "learning_rate": 0.0004274924774322969, "loss": 3.0622, "theoretical_loss": 3.9090717207605246, "tokens_seen": 506920960 }, { "epoch": 0.15, "learning_rate": 0.0004274824473420261, "loss": 3.0204, "theoretical_loss": 3.90901742033728, "tokens_seen": 506986496 }, { "epoch": 0.15, "learning_rate": 0.0004274724172517553, "loss": 2.7641, "theoretical_loss": 3.9089631288978497, "tokens_seen": 507052032 }, { "epoch": 0.15, "learning_rate": 0.00042746238716148444, "loss": 3.313, "theoretical_loss": 3.9089088464395876, "tokens_seen": 507117568 }, { "epoch": 0.15, "learning_rate": 0.0004274523570712137, "loss": 2.8969, "theoretical_loss": 3.908854572959847, "tokens_seen": 507183104 }, { "epoch": 0.15, "learning_rate": 0.0004274423269809428, "loss": 3.2916, "theoretical_loss": 3.9088003084559833, "tokens_seen": 507248640 }, { "epoch": 0.15, "learning_rate": 0.00042743229689067204, "loss": 2.7171, "theoretical_loss": 3.908746052925352, "tokens_seen": 507314176 }, { "epoch": 0.15, "learning_rate": 0.0004274222668004012, "loss": 2.7268, "theoretical_loss": 3.9086918063653115, "tokens_seen": 507379712 }, { "epoch": 0.15, "learning_rate": 0.0004274122367101304, "loss": 3.1031, "theoretical_loss": 3.908637568773219, "tokens_seen": 507445248 }, { "epoch": 0.15, "learning_rate": 0.0004274022066198596, "loss": 2.8652, "theoretical_loss": 3.9085833401464347, "tokens_seen": 507510784 }, { "epoch": 0.15, "learning_rate": 0.00042739217652958876, "loss": 2.9007, "theoretical_loss": 3.908529120482319, "tokens_seen": 507576320 }, { "epoch": 0.15, "learning_rate": 0.00042738214643931794, "loss": 2.8738, "theoretical_loss": 3.908474909778234, "tokens_seen": 507641856 }, { "epoch": 0.15, "learning_rate": 0.0004273721163490472, "loss": 3.0197, "theoretical_loss": 3.908420708031542, "tokens_seen": 507707392 }, { "epoch": 0.15, "learning_rate": 0.0004273620862587763, "loss": 2.8471, "theoretical_loss": 3.9083665152396074, "tokens_seen": 507772928 }, { "epoch": 0.15, "learning_rate": 0.00042735205616850554, "loss": 2.9021, "theoretical_loss": 3.908312331399795, "tokens_seen": 507838464 }, { "epoch": 0.15, "objective/train/docs_used": 298304, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.079557418823242, "objective/train/theoretical_loss": 3.908258156509472, "objective/train/tokens_used": 528364000, "theoretical_loss": 3.908258156509472, "tokens_seen": 507904000 }, { "epoch": 0.15, "learning_rate": 0.00042734202607823467, "loss": 3.1514, "theoretical_loss": 3.908258156509472, "tokens_seen": 507904000 }, { "epoch": 0.15, "learning_rate": 0.0004273319959879639, "loss": 2.9263, "theoretical_loss": 3.9082039905660046, "tokens_seen": 507969536 }, { "epoch": 0.15, "learning_rate": 0.0004273219658976931, "loss": 2.7937, "theoretical_loss": 3.9081498335667613, "tokens_seen": 508035072 }, { "epoch": 0.15, "learning_rate": 0.00042731193580742226, "loss": 3.133, "theoretical_loss": 3.908095685509112, "tokens_seen": 508100608 }, { "epoch": 0.15, "learning_rate": 0.00042730190571715145, "loss": 2.9235, "theoretical_loss": 3.908041546390428, "tokens_seen": 508166144 }, { "epoch": 0.15, "learning_rate": 0.0004272918756268807, "loss": 2.9784, "theoretical_loss": 3.9079874162080794, "tokens_seen": 508231680 }, { "epoch": 0.15, "learning_rate": 0.0004272818455366098, "loss": 3.1766, "theoretical_loss": 3.9079332949594408, "tokens_seen": 508297216 }, { "epoch": 0.15, "learning_rate": 0.00042727181544633904, "loss": 3.2091, "theoretical_loss": 3.907879182641886, "tokens_seen": 508362752 }, { "epoch": 0.15, "learning_rate": 0.00042726178535606817, "loss": 2.9524, "theoretical_loss": 3.907825079252789, "tokens_seen": 508428288 }, { "epoch": 0.15, "learning_rate": 0.0004272517552657974, "loss": 2.9111, "theoretical_loss": 3.9077709847895266, "tokens_seen": 508493824 }, { "epoch": 0.15, "learning_rate": 0.0004272417251755266, "loss": 2.9842, "theoretical_loss": 3.907716899249477, "tokens_seen": 508559360 }, { "epoch": 0.15, "learning_rate": 0.00042723169508525577, "loss": 3.1073, "theoretical_loss": 3.9076628226300167, "tokens_seen": 508624896 }, { "epoch": 0.15, "learning_rate": 0.00042722166499498495, "loss": 2.9411, "theoretical_loss": 3.9076087549285274, "tokens_seen": 508690432 }, { "epoch": 0.15, "learning_rate": 0.00042721163490471413, "loss": 3.1312, "theoretical_loss": 3.907554696142389, "tokens_seen": 508755968 }, { "epoch": 0.15, "learning_rate": 0.0004272016048144433, "loss": 3.1056, "theoretical_loss": 3.9075006462689825, "tokens_seen": 508821504 }, { "epoch": 0.15, "learning_rate": 0.00042719157472417255, "loss": 3.0852, "theoretical_loss": 3.9074466053056915, "tokens_seen": 508887040 }, { "epoch": 0.15, "learning_rate": 0.0004271815446339017, "loss": 3.1324, "theoretical_loss": 3.9073925732498997, "tokens_seen": 508952576 }, { "epoch": 0.15, "learning_rate": 0.0004271715145436309, "loss": 2.8519, "theoretical_loss": 3.907338550098993, "tokens_seen": 509018112 }, { "epoch": 0.15, "learning_rate": 0.0004271614844533601, "loss": 3.0767, "theoretical_loss": 3.907284535850356, "tokens_seen": 509083648 }, { "epoch": 0.15, "learning_rate": 0.00042715145436308927, "loss": 3.1534, "theoretical_loss": 3.9072305305013773, "tokens_seen": 509149184 }, { "epoch": 0.15, "learning_rate": 0.0004271414242728185, "loss": 2.9208, "theoretical_loss": 3.9071765340494453, "tokens_seen": 509214720 }, { "epoch": 0.15, "learning_rate": 0.00042713139418254763, "loss": 3.0471, "theoretical_loss": 3.9071225464919483, "tokens_seen": 509280256 }, { "epoch": 0.15, "learning_rate": 0.00042712136409227687, "loss": 2.9369, "theoretical_loss": 3.907068567826278, "tokens_seen": 509345792 }, { "epoch": 0.15, "learning_rate": 0.00042711133400200605, "loss": 3.0517, "theoretical_loss": 3.9070145980498263, "tokens_seen": 509411328 }, { "epoch": 0.15, "learning_rate": 0.00042710130391173523, "loss": 2.983, "theoretical_loss": 3.906960637159985, "tokens_seen": 509476864 }, { "epoch": 0.15, "objective/train/docs_used": 298986, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2158470153808594, "objective/train/theoretical_loss": 3.9069066851541487, "objective/train/tokens_used": 530002400, "theoretical_loss": 3.9069066851541487, "tokens_seen": 509542400 }, { "epoch": 0.15, "learning_rate": 0.0004270912738214644, "loss": 3.0958, "theoretical_loss": 3.9069066851541487, "tokens_seen": 509542400 }, { "epoch": 0.15, "learning_rate": 0.0004270812437311936, "loss": 2.8435, "theoretical_loss": 3.906852742029712, "tokens_seen": 509607936 }, { "epoch": 0.15, "learning_rate": 0.0004270712136409228, "loss": 3.1957, "theoretical_loss": 3.906798807784072, "tokens_seen": 509673472 }, { "epoch": 0.15, "learning_rate": 0.000427061183550652, "loss": 3.0716, "theoretical_loss": 3.9067448824146243, "tokens_seen": 509739008 }, { "epoch": 0.15, "learning_rate": 0.00042705115346038114, "loss": 2.9525, "theoretical_loss": 3.9066909659187683, "tokens_seen": 509804544 }, { "epoch": 0.15, "learning_rate": 0.00042704112337011037, "loss": 2.8851, "theoretical_loss": 3.906637058293903, "tokens_seen": 509870080 }, { "epoch": 0.15, "learning_rate": 0.0004270310932798395, "loss": 3.1575, "theoretical_loss": 3.906583159537429, "tokens_seen": 509935616 }, { "epoch": 0.15, "learning_rate": 0.00042702106318956873, "loss": 2.9141, "theoretical_loss": 3.9065292696467475, "tokens_seen": 510001152 }, { "epoch": 0.15, "learning_rate": 0.0004270110330992979, "loss": 3.2035, "theoretical_loss": 3.906475388619262, "tokens_seen": 510066688 }, { "epoch": 0.15, "learning_rate": 0.0004270010030090271, "loss": 2.7468, "theoretical_loss": 3.906421516452376, "tokens_seen": 510132224 }, { "epoch": 0.15, "learning_rate": 0.0004269909729187563, "loss": 3.0385, "theoretical_loss": 3.9063676531434934, "tokens_seen": 510197760 }, { "epoch": 0.15, "learning_rate": 0.0004269809428284855, "loss": 2.796, "theoretical_loss": 3.9063137986900216, "tokens_seen": 510263296 }, { "epoch": 0.15, "learning_rate": 0.00042697091273821464, "loss": 3.0383, "theoretical_loss": 3.906259953089367, "tokens_seen": 510328832 }, { "epoch": 0.15, "learning_rate": 0.0004269608826479439, "loss": 2.6359, "theoretical_loss": 3.9062061163389377, "tokens_seen": 510394368 }, { "epoch": 0.15, "learning_rate": 0.000426950852557673, "loss": 2.7839, "theoretical_loss": 3.906152288436143, "tokens_seen": 510459904 }, { "epoch": 0.15, "learning_rate": 0.00042694082246740224, "loss": 2.876, "theoretical_loss": 3.906098469378393, "tokens_seen": 510525440 }, { "epoch": 0.15, "learning_rate": 0.0004269307923771314, "loss": 2.8037, "theoretical_loss": 3.9060446591630997, "tokens_seen": 510590976 }, { "epoch": 0.15, "learning_rate": 0.0004269207622868606, "loss": 2.8069, "theoretical_loss": 3.905990857787675, "tokens_seen": 510656512 }, { "epoch": 0.15, "learning_rate": 0.0004269107321965898, "loss": 2.7768, "theoretical_loss": 3.9059370652495327, "tokens_seen": 510722048 }, { "epoch": 0.15, "learning_rate": 0.00042690070210631896, "loss": 3.1726, "theoretical_loss": 3.905883281546088, "tokens_seen": 510787584 }, { "epoch": 0.15, "learning_rate": 0.00042689067201604814, "loss": 3.0302, "theoretical_loss": 3.9058295066747553, "tokens_seen": 510853120 }, { "epoch": 0.15, "learning_rate": 0.0004268806419257774, "loss": 3.133, "theoretical_loss": 3.9057757406329534, "tokens_seen": 510918656 }, { "epoch": 0.15, "learning_rate": 0.0004268706118355065, "loss": 3.2004, "theoretical_loss": 3.9057219834180987, "tokens_seen": 510984192 }, { "epoch": 0.15, "learning_rate": 0.00042686058174523574, "loss": 2.9723, "theoretical_loss": 3.9056682350276106, "tokens_seen": 511049728 }, { "epoch": 0.15, "learning_rate": 0.00042685055165496487, "loss": 3.0879, "theoretical_loss": 3.90561449545891, "tokens_seen": 511115264 }, { "epoch": 0.15, "objective/train/docs_used": 299996, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.871511697769165, "objective/train/theoretical_loss": 3.905560764709417, "objective/train/tokens_used": 531640800, "theoretical_loss": 3.905560764709417, "tokens_seen": 511180800 }, { "epoch": 0.15, "learning_rate": 0.0004268405215646941, "loss": 2.808, "theoretical_loss": 3.905560764709417, "tokens_seen": 511180800 }, { "epoch": 0.15, "learning_rate": 0.0004268304914744233, "loss": 3.1141, "theoretical_loss": 3.905507042776554, "tokens_seen": 511246336 }, { "epoch": 0.15, "learning_rate": 0.00042682046138415246, "loss": 3.089, "theoretical_loss": 3.905453329657745, "tokens_seen": 511311872 }, { "epoch": 0.15, "learning_rate": 0.00042681043129388165, "loss": 3.1051, "theoretical_loss": 3.905399625350414, "tokens_seen": 511377408 }, { "epoch": 0.15, "learning_rate": 0.0004268004012036109, "loss": 2.9046, "theoretical_loss": 3.9053459298519875, "tokens_seen": 511442944 }, { "epoch": 0.16, "learning_rate": 0.00042679037111334, "loss": 2.8459, "theoretical_loss": 3.905292243159891, "tokens_seen": 511508480 }, { "epoch": 0.16, "learning_rate": 0.00042678034102306924, "loss": 2.867, "theoretical_loss": 3.905238565271553, "tokens_seen": 511574016 }, { "epoch": 0.16, "learning_rate": 0.00042677031093279837, "loss": 2.9912, "theoretical_loss": 3.9051848961844007, "tokens_seen": 511639552 }, { "epoch": 0.16, "learning_rate": 0.0004267602808425276, "loss": 3.0169, "theoretical_loss": 3.9051312358958663, "tokens_seen": 511705088 }, { "epoch": 0.16, "learning_rate": 0.0004267502507522568, "loss": 2.7919, "theoretical_loss": 3.9050775844033785, "tokens_seen": 511770624 }, { "epoch": 0.16, "learning_rate": 0.00042674022066198597, "loss": 3.0171, "theoretical_loss": 3.905023941704371, "tokens_seen": 511836160 }, { "epoch": 0.16, "learning_rate": 0.00042673019057171515, "loss": 2.8172, "theoretical_loss": 3.904970307796276, "tokens_seen": 511901696 }, { "epoch": 0.16, "learning_rate": 0.00042672016048144433, "loss": 2.803, "theoretical_loss": 3.9049166826765283, "tokens_seen": 511967232 }, { "epoch": 0.16, "learning_rate": 0.0004267101303911735, "loss": 3.141, "theoretical_loss": 3.9048630663425623, "tokens_seen": 512032768 }, { "epoch": 0.16, "learning_rate": 0.00042670010030090275, "loss": 3.0635, "theoretical_loss": 3.9048094587918154, "tokens_seen": 512098304 }, { "epoch": 0.16, "learning_rate": 0.0004266900702106319, "loss": 2.8628, "theoretical_loss": 3.904755860021724, "tokens_seen": 512163840 }, { "epoch": 0.16, "learning_rate": 0.0004266800401203611, "loss": 2.7707, "theoretical_loss": 3.904702270029727, "tokens_seen": 512229376 }, { "epoch": 0.16, "learning_rate": 0.00042667001003009024, "loss": 2.894, "theoretical_loss": 3.904648688813264, "tokens_seen": 512294912 }, { "epoch": 0.16, "learning_rate": 0.00042665997993981947, "loss": 2.9793, "theoretical_loss": 3.9045951163697756, "tokens_seen": 512360448 }, { "epoch": 0.16, "learning_rate": 0.00042664994984954865, "loss": 3.0136, "theoretical_loss": 3.9045415526967036, "tokens_seen": 512425984 }, { "epoch": 0.16, "learning_rate": 0.00042663991975927783, "loss": 3.1039, "theoretical_loss": 3.904487997791491, "tokens_seen": 512491520 }, { "epoch": 0.16, "learning_rate": 0.000426629889669007, "loss": 2.6166, "theoretical_loss": 3.904434451651581, "tokens_seen": 512557056 }, { "epoch": 0.16, "learning_rate": 0.00042661985957873625, "loss": 2.8251, "theoretical_loss": 3.904380914274419, "tokens_seen": 512622592 }, { "epoch": 0.16, "learning_rate": 0.0004266098294884654, "loss": 2.9569, "theoretical_loss": 3.9043273856574507, "tokens_seen": 512688128 }, { "epoch": 0.16, "learning_rate": 0.0004265997993981946, "loss": 2.8449, "theoretical_loss": 3.9042738657981237, "tokens_seen": 512753664 }, { "epoch": 0.16, "objective/train/docs_used": 300643, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9767351150512695, "objective/train/theoretical_loss": 3.9042203546938854, "objective/train/tokens_used": 533279200, "theoretical_loss": 3.9042203546938854, "tokens_seen": 512819200 }, { "epoch": 0.16, "learning_rate": 0.00042658976930792374, "loss": 2.9345, "theoretical_loss": 3.9042203546938854, "tokens_seen": 512819200 }, { "epoch": 0.16, "learning_rate": 0.000426579739217653, "loss": 2.6117, "theoretical_loss": 3.9041668523421853, "tokens_seen": 512884736 }, { "epoch": 0.16, "learning_rate": 0.00042656970912738216, "loss": 2.9333, "theoretical_loss": 3.9041133587404744, "tokens_seen": 512950272 }, { "epoch": 0.16, "learning_rate": 0.00042655967903711134, "loss": 3.0242, "theoretical_loss": 3.9040598738862027, "tokens_seen": 513015808 }, { "epoch": 0.16, "learning_rate": 0.0004265496489468405, "loss": 3.1475, "theoretical_loss": 3.904006397776823, "tokens_seen": 513081344 }, { "epoch": 0.16, "learning_rate": 0.0004265396188565697, "loss": 3.0377, "theoretical_loss": 3.90395293040979, "tokens_seen": 513146880 }, { "epoch": 0.16, "learning_rate": 0.0004265295887662989, "loss": 3.0351, "theoretical_loss": 3.903899471782557, "tokens_seen": 513212416 }, { "epoch": 0.16, "learning_rate": 0.0004265195586760281, "loss": 3.0313, "theoretical_loss": 3.90384602189258, "tokens_seen": 513277952 }, { "epoch": 0.16, "learning_rate": 0.00042650952858575724, "loss": 2.949, "theoretical_loss": 3.9037925807373153, "tokens_seen": 513343488 }, { "epoch": 0.16, "learning_rate": 0.0004264994984954865, "loss": 3.159, "theoretical_loss": 3.9037391483142208, "tokens_seen": 513409024 }, { "epoch": 0.16, "learning_rate": 0.0004264894684052156, "loss": 3.0819, "theoretical_loss": 3.9036857246207557, "tokens_seen": 513474560 }, { "epoch": 0.16, "learning_rate": 0.00042647943831494484, "loss": 2.8282, "theoretical_loss": 3.9036323096543795, "tokens_seen": 513540096 }, { "epoch": 0.16, "learning_rate": 0.000426469408224674, "loss": 3.0349, "theoretical_loss": 3.903578903412553, "tokens_seen": 513605632 }, { "epoch": 0.16, "learning_rate": 0.0004264593781344032, "loss": 2.7911, "theoretical_loss": 3.9035255058927394, "tokens_seen": 513671168 }, { "epoch": 0.16, "learning_rate": 0.0004264493480441324, "loss": 3.1141, "theoretical_loss": 3.9034721170924005, "tokens_seen": 513736704 }, { "epoch": 0.16, "learning_rate": 0.0004264393179538616, "loss": 2.9924, "theoretical_loss": 3.903418737009, "tokens_seen": 513802240 }, { "epoch": 0.16, "learning_rate": 0.00042642928786359075, "loss": 2.9924, "theoretical_loss": 3.9033653656400045, "tokens_seen": 513867776 }, { "epoch": 0.16, "learning_rate": 0.00042641925777332, "loss": 2.8468, "theoretical_loss": 3.903312002982879, "tokens_seen": 513933312 }, { "epoch": 0.16, "learning_rate": 0.00042640922768304916, "loss": 2.8712, "theoretical_loss": 3.9032586490350916, "tokens_seen": 513998848 }, { "epoch": 0.16, "learning_rate": 0.00042639919759277834, "loss": 3.0535, "theoretical_loss": 3.9032053037941106, "tokens_seen": 514064384 }, { "epoch": 0.16, "learning_rate": 0.0004263891675025076, "loss": 3.075, "theoretical_loss": 3.9031519672574047, "tokens_seen": 514129920 }, { "epoch": 0.16, "learning_rate": 0.0004263791374122367, "loss": 2.8464, "theoretical_loss": 3.903098639422445, "tokens_seen": 514195456 }, { "epoch": 0.16, "learning_rate": 0.00042636910732196594, "loss": 2.9903, "theoretical_loss": 3.9030453202867035, "tokens_seen": 514260992 }, { "epoch": 0.16, "learning_rate": 0.00042635907723169507, "loss": 2.8887, "theoretical_loss": 3.902992009847652, "tokens_seen": 514326528 }, { "epoch": 0.16, "learning_rate": 0.0004263490471414243, "loss": 2.8372, "theoretical_loss": 3.9029387081027638, "tokens_seen": 514392064 }, { "epoch": 0.16, "objective/train/docs_used": 301818, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.592280864715576, "objective/train/theoretical_loss": 3.9028854150495143, "objective/train/tokens_used": 534917600, "theoretical_loss": 3.9028854150495143, "tokens_seen": 514457600 }, { "epoch": 0.16, "learning_rate": 0.0004263390170511535, "loss": 2.7809, "theoretical_loss": 3.9028854150495143, "tokens_seen": 514457600 }, { "epoch": 0.16, "learning_rate": 0.00042632898696088266, "loss": 2.8517, "theoretical_loss": 3.9028321306853795, "tokens_seen": 514523136 }, { "epoch": 0.16, "learning_rate": 0.00042631895687061185, "loss": 2.9973, "theoretical_loss": 3.9027788550078357, "tokens_seen": 514588672 }, { "epoch": 0.16, "learning_rate": 0.0004263089267803411, "loss": 2.901, "theoretical_loss": 3.902725588014361, "tokens_seen": 514654208 }, { "epoch": 0.16, "learning_rate": 0.0004262988966900702, "loss": 3.0018, "theoretical_loss": 3.9026723297024337, "tokens_seen": 514719744 }, { "epoch": 0.16, "learning_rate": 0.00042628886659979944, "loss": 2.9766, "theoretical_loss": 3.9026190800695346, "tokens_seen": 514785280 }, { "epoch": 0.16, "learning_rate": 0.00042627883650952857, "loss": 2.8017, "theoretical_loss": 3.9025658391131444, "tokens_seen": 514850816 }, { "epoch": 0.16, "learning_rate": 0.0004262688064192578, "loss": 2.7732, "theoretical_loss": 3.902512606830745, "tokens_seen": 514916352 }, { "epoch": 0.16, "learning_rate": 0.000426258776328987, "loss": 2.7042, "theoretical_loss": 3.9024593832198198, "tokens_seen": 514981888 }, { "epoch": 0.16, "learning_rate": 0.00042624874623871617, "loss": 2.6781, "theoretical_loss": 3.902406168277853, "tokens_seen": 515047424 }, { "epoch": 0.16, "learning_rate": 0.00042623871614844535, "loss": 2.9133, "theoretical_loss": 3.9023529620023294, "tokens_seen": 515112960 }, { "epoch": 0.16, "learning_rate": 0.00042622868605817453, "loss": 2.7778, "theoretical_loss": 3.902299764390736, "tokens_seen": 515178496 }, { "epoch": 0.16, "learning_rate": 0.0004262186559679037, "loss": 3.1145, "theoretical_loss": 3.902246575440559, "tokens_seen": 515244032 }, { "epoch": 0.16, "learning_rate": 0.00042620862587763295, "loss": 2.8735, "theoretical_loss": 3.902193395149288, "tokens_seen": 515309568 }, { "epoch": 0.16, "learning_rate": 0.0004261985957873621, "loss": 2.6957, "theoretical_loss": 3.9021402235144116, "tokens_seen": 515375104 }, { "epoch": 0.16, "learning_rate": 0.0004261885656970913, "loss": 2.7606, "theoretical_loss": 3.9020870605334204, "tokens_seen": 515440640 }, { "epoch": 0.16, "learning_rate": 0.00042617853560682044, "loss": 2.9639, "theoretical_loss": 3.9020339062038065, "tokens_seen": 515506176 }, { "epoch": 0.16, "learning_rate": 0.00042616850551654967, "loss": 2.9819, "theoretical_loss": 3.9019807605230614, "tokens_seen": 515571712 }, { "epoch": 0.16, "learning_rate": 0.00042615847542627885, "loss": 2.827, "theoretical_loss": 3.90192762348868, "tokens_seen": 515637248 }, { "epoch": 0.16, "learning_rate": 0.00042614844533600803, "loss": 3.0162, "theoretical_loss": 3.9018744950981556, "tokens_seen": 515702784 }, { "epoch": 0.16, "learning_rate": 0.0004261384152457372, "loss": 3.0663, "theoretical_loss": 3.901821375348985, "tokens_seen": 515768320 }, { "epoch": 0.16, "learning_rate": 0.00042612838515546645, "loss": 2.9748, "theoretical_loss": 3.9017682642386644, "tokens_seen": 515833856 }, { "epoch": 0.16, "learning_rate": 0.0004261183550651956, "loss": 3.0898, "theoretical_loss": 3.9017151617646917, "tokens_seen": 515899392 }, { "epoch": 0.16, "learning_rate": 0.0004261083249749248, "loss": 2.8773, "theoretical_loss": 3.901662067924566, "tokens_seen": 515964928 }, { "epoch": 0.16, "learning_rate": 0.00042609829488465394, "loss": 2.8027, "theoretical_loss": 3.9016089827157865, "tokens_seen": 516030464 }, { "epoch": 0.16, "objective/train/docs_used": 302480, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9364280700683594, "objective/train/theoretical_loss": 3.9015559061358545, "objective/train/tokens_used": 536556000, "theoretical_loss": 3.9015559061358545, "tokens_seen": 516096000 }, { "epoch": 0.16, "learning_rate": 0.0004260882647943832, "loss": 2.9585, "theoretical_loss": 3.9015559061358545, "tokens_seen": 516096000 }, { "epoch": 0.16, "learning_rate": 0.00042607823470411236, "loss": 2.6055, "theoretical_loss": 3.901502838182272, "tokens_seen": 516161536 }, { "epoch": 0.16, "learning_rate": 0.00042606820461384154, "loss": 3.1029, "theoretical_loss": 3.901449778852542, "tokens_seen": 516227072 }, { "epoch": 0.16, "learning_rate": 0.0004260581745235707, "loss": 3.0751, "theoretical_loss": 3.9013967281441677, "tokens_seen": 516292608 }, { "epoch": 0.16, "learning_rate": 0.0004260481444332999, "loss": 3.002, "theoretical_loss": 3.9013436860546555, "tokens_seen": 516358144 }, { "epoch": 0.16, "learning_rate": 0.0004260381143430291, "loss": 2.8306, "theoretical_loss": 3.901290652581511, "tokens_seen": 516423680 }, { "epoch": 0.16, "learning_rate": 0.0004260280842527583, "loss": 2.888, "theoretical_loss": 3.901237627722241, "tokens_seen": 516489216 }, { "epoch": 0.16, "learning_rate": 0.00042601805416248744, "loss": 2.8373, "theoretical_loss": 3.9011846114743536, "tokens_seen": 516554752 }, { "epoch": 0.16, "learning_rate": 0.0004260080240722167, "loss": 2.7766, "theoretical_loss": 3.901131603835359, "tokens_seen": 516620288 }, { "epoch": 0.16, "learning_rate": 0.0004259979939819458, "loss": 3.003, "theoretical_loss": 3.9010786048027666, "tokens_seen": 516685824 }, { "epoch": 0.16, "learning_rate": 0.00042598796389167504, "loss": 2.819, "theoretical_loss": 3.9010256143740873, "tokens_seen": 516751360 }, { "epoch": 0.16, "learning_rate": 0.0004259779338014042, "loss": 3.0103, "theoretical_loss": 3.900972632546835, "tokens_seen": 516816896 }, { "epoch": 0.16, "learning_rate": 0.0004259679037111334, "loss": 3.0061, "theoretical_loss": 3.900919659318521, "tokens_seen": 516882432 }, { "epoch": 0.16, "learning_rate": 0.0004259578736208626, "loss": 3.2129, "theoretical_loss": 3.9008666946866613, "tokens_seen": 516947968 }, { "epoch": 0.16, "learning_rate": 0.0004259478435305918, "loss": 3.0836, "theoretical_loss": 3.900813738648771, "tokens_seen": 517013504 }, { "epoch": 0.16, "learning_rate": 0.00042593781344032095, "loss": 2.9082, "theoretical_loss": 3.900760791202366, "tokens_seen": 517079040 }, { "epoch": 0.16, "learning_rate": 0.0004259277833500502, "loss": 2.9466, "theoretical_loss": 3.900707852344964, "tokens_seen": 517144576 }, { "epoch": 0.16, "learning_rate": 0.0004259177532597793, "loss": 3.081, "theoretical_loss": 3.9006549220740836, "tokens_seen": 517210112 }, { "epoch": 0.16, "learning_rate": 0.00042590772316950854, "loss": 2.8449, "theoretical_loss": 3.9006020003872446, "tokens_seen": 517275648 }, { "epoch": 0.16, "learning_rate": 0.0004258976930792377, "loss": 2.8311, "theoretical_loss": 3.9005490872819673, "tokens_seen": 517341184 }, { "epoch": 0.16, "learning_rate": 0.0004258876629889669, "loss": 2.991, "theoretical_loss": 3.900496182755773, "tokens_seen": 517406720 }, { "epoch": 0.16, "learning_rate": 0.0004258776328986961, "loss": 2.9178, "theoretical_loss": 3.9004432868061856, "tokens_seen": 517472256 }, { "epoch": 0.16, "learning_rate": 0.00042586760280842527, "loss": 2.7577, "theoretical_loss": 3.9003903994307274, "tokens_seen": 517537792 }, { "epoch": 0.16, "learning_rate": 0.00042585757271815445, "loss": 2.7745, "theoretical_loss": 3.9003375206269233, "tokens_seen": 517603328 }, { "epoch": 0.16, "learning_rate": 0.0004258475426278837, "loss": 2.9678, "theoretical_loss": 3.9002846503923, "tokens_seen": 517668864 }, { "epoch": 0.16, "objective/train/docs_used": 303675, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.994904041290283, "objective/train/theoretical_loss": 3.9002317887243834, "objective/train/tokens_used": 538194400, "theoretical_loss": 3.9002317887243834, "tokens_seen": 517734400 }, { "epoch": 0.16, "learning_rate": 0.0004258375125376128, "loss": 2.8722, "theoretical_loss": 3.9002317887243834, "tokens_seen": 517734400 }, { "epoch": 0.16, "learning_rate": 0.00042582748244734205, "loss": 2.9304, "theoretical_loss": 3.9001789356207013, "tokens_seen": 517799936 }, { "epoch": 0.16, "learning_rate": 0.0004258174523570712, "loss": 2.8775, "theoretical_loss": 3.900126091078783, "tokens_seen": 517865472 }, { "epoch": 0.16, "learning_rate": 0.0004258074222668004, "loss": 3.0728, "theoretical_loss": 3.9000732550961574, "tokens_seen": 517931008 }, { "epoch": 0.16, "learning_rate": 0.0004257973921765296, "loss": 2.9802, "theoretical_loss": 3.9000204276703565, "tokens_seen": 517996544 }, { "epoch": 0.16, "learning_rate": 0.00042578736208625877, "loss": 2.889, "theoretical_loss": 3.8999676087989115, "tokens_seen": 518062080 }, { "epoch": 0.16, "learning_rate": 0.00042577733199598795, "loss": 3.0984, "theoretical_loss": 3.8999147984793554, "tokens_seen": 518127616 }, { "epoch": 0.16, "learning_rate": 0.0004257673019057172, "loss": 2.8683, "theoretical_loss": 3.8998619967092223, "tokens_seen": 518193152 }, { "epoch": 0.16, "learning_rate": 0.0004257572718154463, "loss": 2.7482, "theoretical_loss": 3.8998092034860474, "tokens_seen": 518258688 }, { "epoch": 0.16, "learning_rate": 0.00042574724172517555, "loss": 3.1747, "theoretical_loss": 3.8997564188073657, "tokens_seen": 518324224 }, { "epoch": 0.16, "learning_rate": 0.0004257372116349047, "loss": 2.7086, "theoretical_loss": 3.8997036426707155, "tokens_seen": 518389760 }, { "epoch": 0.16, "learning_rate": 0.0004257271815446339, "loss": 2.925, "theoretical_loss": 3.8996508750736343, "tokens_seen": 518455296 }, { "epoch": 0.16, "learning_rate": 0.0004257171514543631, "loss": 2.9987, "theoretical_loss": 3.89959811601366, "tokens_seen": 518520832 }, { "epoch": 0.16, "learning_rate": 0.0004257071213640923, "loss": 2.7581, "theoretical_loss": 3.8995453654883345, "tokens_seen": 518586368 }, { "epoch": 0.16, "learning_rate": 0.00042569709127382145, "loss": 2.8995, "theoretical_loss": 3.899492623495198, "tokens_seen": 518651904 }, { "epoch": 0.16, "learning_rate": 0.00042568706118355064, "loss": 2.8678, "theoretical_loss": 3.8994398900317924, "tokens_seen": 518717440 }, { "epoch": 0.16, "learning_rate": 0.0004256770310932798, "loss": 2.7447, "theoretical_loss": 3.8993871650956606, "tokens_seen": 518782976 }, { "epoch": 0.16, "learning_rate": 0.00042566700100300905, "loss": 2.8438, "theoretical_loss": 3.899334448684348, "tokens_seen": 518848512 }, { "epoch": 0.16, "learning_rate": 0.00042565697091273823, "loss": 2.8911, "theoretical_loss": 3.8992817407953986, "tokens_seen": 518914048 }, { "epoch": 0.16, "learning_rate": 0.0004256469408224674, "loss": 2.7872, "theoretical_loss": 3.8992290414263593, "tokens_seen": 518979584 }, { "epoch": 0.16, "learning_rate": 0.00042563691073219665, "loss": 2.8124, "theoretical_loss": 3.8991763505747765, "tokens_seen": 519045120 }, { "epoch": 0.16, "learning_rate": 0.0004256268806419258, "loss": 2.9019, "theoretical_loss": 3.8991236682381993, "tokens_seen": 519110656 }, { "epoch": 0.16, "learning_rate": 0.000425616850551655, "loss": 2.8403, "theoretical_loss": 3.899070994414176, "tokens_seen": 519176192 }, { "epoch": 0.16, "learning_rate": 0.00042560682046138414, "loss": 2.9856, "theoretical_loss": 3.899018329100257, "tokens_seen": 519241728 }, { "epoch": 0.16, "learning_rate": 0.0004255967903711134, "loss": 3.0539, "theoretical_loss": 3.8989656722939943, "tokens_seen": 519307264 }, { "epoch": 0.16, "objective/train/docs_used": 304298, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.55254864692688, "objective/train/theoretical_loss": 3.8989130239929395, "objective/train/tokens_used": 539832800, "theoretical_loss": 3.8989130239929395, "tokens_seen": 519372800 }, { "epoch": 0.16, "learning_rate": 0.00042558676028084256, "loss": 2.7837, "theoretical_loss": 3.8989130239929395, "tokens_seen": 519372800 }, { "epoch": 0.16, "learning_rate": 0.00042557673019057174, "loss": 3.1594, "theoretical_loss": 3.8988603841946463, "tokens_seen": 519438336 }, { "epoch": 0.16, "learning_rate": 0.0004255667001003009, "loss": 3.1587, "theoretical_loss": 3.8988077528966687, "tokens_seen": 519503872 }, { "epoch": 0.16, "learning_rate": 0.0004255566700100301, "loss": 2.894, "theoretical_loss": 3.898755130096562, "tokens_seen": 519569408 }, { "epoch": 0.16, "learning_rate": 0.0004255466399197593, "loss": 2.8944, "theoretical_loss": 3.898702515791882, "tokens_seen": 519634944 }, { "epoch": 0.16, "learning_rate": 0.0004255366098294885, "loss": 2.966, "theoretical_loss": 3.8986499099801875, "tokens_seen": 519700480 }, { "epoch": 0.16, "learning_rate": 0.00042552657973921764, "loss": 3.0096, "theoretical_loss": 3.898597312659035, "tokens_seen": 519766016 }, { "epoch": 0.16, "learning_rate": 0.0004255165496489469, "loss": 3.0903, "theoretical_loss": 3.8985447238259856, "tokens_seen": 519831552 }, { "epoch": 0.16, "learning_rate": 0.000425506519558676, "loss": 2.9664, "theoretical_loss": 3.8984921434785984, "tokens_seen": 519897088 }, { "epoch": 0.16, "learning_rate": 0.00042549648946840524, "loss": 2.7405, "theoretical_loss": 3.8984395716144347, "tokens_seen": 519962624 }, { "epoch": 0.16, "learning_rate": 0.0004254864593781344, "loss": 3.0711, "theoretical_loss": 3.898387008231058, "tokens_seen": 520028160 }, { "epoch": 0.16, "learning_rate": 0.0004254764292878636, "loss": 3.0678, "theoretical_loss": 3.8983344533260307, "tokens_seen": 520093696 }, { "epoch": 0.16, "learning_rate": 0.0004254663991975928, "loss": 2.9319, "theoretical_loss": 3.8982819068969174, "tokens_seen": 520159232 }, { "epoch": 0.16, "learning_rate": 0.000425456369107322, "loss": 2.8959, "theoretical_loss": 3.8982293689412835, "tokens_seen": 520224768 }, { "epoch": 0.16, "learning_rate": 0.00042544633901705115, "loss": 2.8478, "theoretical_loss": 3.8981768394566956, "tokens_seen": 520290304 }, { "epoch": 0.16, "learning_rate": 0.0004254363089267804, "loss": 3.0487, "theoretical_loss": 3.898124318440721, "tokens_seen": 520355840 }, { "epoch": 0.16, "learning_rate": 0.0004254262788365095, "loss": 3.0492, "theoretical_loss": 3.8980718058909276, "tokens_seen": 520421376 }, { "epoch": 0.16, "learning_rate": 0.00042541624874623874, "loss": 2.9961, "theoretical_loss": 3.8980193018048857, "tokens_seen": 520486912 }, { "epoch": 0.16, "learning_rate": 0.0004254062186559679, "loss": 3.0865, "theoretical_loss": 3.8979668061801656, "tokens_seen": 520552448 }, { "epoch": 0.16, "learning_rate": 0.0004253961885656971, "loss": 2.9966, "theoretical_loss": 3.8979143190143377, "tokens_seen": 520617984 }, { "epoch": 0.16, "learning_rate": 0.0004253861584754263, "loss": 2.7913, "theoretical_loss": 3.8978618403049747, "tokens_seen": 520683520 }, { "epoch": 0.16, "learning_rate": 0.00042537612838515547, "loss": 3.017, "theoretical_loss": 3.8978093700496514, "tokens_seen": 520749056 }, { "epoch": 0.16, "learning_rate": 0.00042536609829488465, "loss": 2.8749, "theoretical_loss": 3.897756908245941, "tokens_seen": 520814592 }, { "epoch": 0.16, "learning_rate": 0.0004253560682046139, "loss": 2.9132, "theoretical_loss": 3.897704454891419, "tokens_seen": 520880128 }, { "epoch": 0.16, "learning_rate": 0.000425346038114343, "loss": 2.6791, "theoretical_loss": 3.897652009983662, "tokens_seen": 520945664 }, { "epoch": 0.16, "objective/train/docs_used": 305033, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.236427068710327, "objective/train/theoretical_loss": 3.897599573520247, "objective/train/tokens_used": 541471200, "theoretical_loss": 3.897599573520247, "tokens_seen": 521011200 }, { "epoch": 0.16, "learning_rate": 0.00042533600802407225, "loss": 3.0652, "theoretical_loss": 3.897599573520247, "tokens_seen": 521011200 }, { "epoch": 0.16, "learning_rate": 0.0004253259779338014, "loss": 2.7909, "theoretical_loss": 3.897547145498753, "tokens_seen": 521076736 }, { "epoch": 0.16, "learning_rate": 0.0004253159478435306, "loss": 2.945, "theoretical_loss": 3.8974947259167596, "tokens_seen": 521142272 }, { "epoch": 0.16, "learning_rate": 0.0004253059177532598, "loss": 2.9162, "theoretical_loss": 3.8974423147718467, "tokens_seen": 521207808 }, { "epoch": 0.16, "learning_rate": 0.00042529588766298897, "loss": 3.0102, "theoretical_loss": 3.8973899120615956, "tokens_seen": 521273344 }, { "epoch": 0.16, "learning_rate": 0.00042528585757271815, "loss": 2.8023, "theoretical_loss": 3.8973375177835896, "tokens_seen": 521338880 }, { "epoch": 0.16, "learning_rate": 0.0004252758274824474, "loss": 3.0294, "theoretical_loss": 3.897285131935411, "tokens_seen": 521404416 }, { "epoch": 0.16, "learning_rate": 0.0004252657973921765, "loss": 3.1221, "theoretical_loss": 3.8972327545146443, "tokens_seen": 521469952 }, { "epoch": 0.16, "learning_rate": 0.00042525576730190575, "loss": 2.6076, "theoretical_loss": 3.897180385518876, "tokens_seen": 521535488 }, { "epoch": 0.16, "learning_rate": 0.0004252457372116349, "loss": 3.0293, "theoretical_loss": 3.897128024945692, "tokens_seen": 521601024 }, { "epoch": 0.16, "learning_rate": 0.0004252357071213641, "loss": 2.9325, "theoretical_loss": 3.8970756727926785, "tokens_seen": 521666560 }, { "epoch": 0.16, "learning_rate": 0.0004252256770310933, "loss": 3.0425, "theoretical_loss": 3.8970233290574257, "tokens_seen": 521732096 }, { "epoch": 0.16, "learning_rate": 0.0004252156469408225, "loss": 2.8348, "theoretical_loss": 3.896970993737522, "tokens_seen": 521797632 }, { "epoch": 0.16, "learning_rate": 0.00042520561685055165, "loss": 3.0828, "theoretical_loss": 3.896918666830558, "tokens_seen": 521863168 }, { "epoch": 0.16, "learning_rate": 0.00042519558676028084, "loss": 2.8449, "theoretical_loss": 3.8968663483341253, "tokens_seen": 521928704 }, { "epoch": 0.16, "learning_rate": 0.00042518555667001, "loss": 2.8192, "theoretical_loss": 3.8968140382458154, "tokens_seen": 521994240 }, { "epoch": 0.16, "learning_rate": 0.00042517552657973925, "loss": 2.9585, "theoretical_loss": 3.8967617365632217, "tokens_seen": 522059776 }, { "epoch": 0.16, "learning_rate": 0.0004251654964894684, "loss": 2.8681, "theoretical_loss": 3.8967094432839398, "tokens_seen": 522125312 }, { "epoch": 0.16, "learning_rate": 0.0004251554663991976, "loss": 2.9713, "theoretical_loss": 3.8966571584055645, "tokens_seen": 522190848 }, { "epoch": 0.16, "learning_rate": 0.0004251454363089268, "loss": 2.908, "theoretical_loss": 3.896604881925691, "tokens_seen": 522256384 }, { "epoch": 0.16, "learning_rate": 0.000425135406218656, "loss": 2.7829, "theoretical_loss": 3.8965526138419184, "tokens_seen": 522321920 }, { "epoch": 0.16, "learning_rate": 0.00042512537612838516, "loss": 2.8642, "theoretical_loss": 3.8965003541518435, "tokens_seen": 522387456 }, { "epoch": 0.16, "learning_rate": 0.00042511534603811434, "loss": 2.9174, "theoretical_loss": 3.896448102853067, "tokens_seen": 522452992 }, { "epoch": 0.16, "learning_rate": 0.0004251053159478435, "loss": 2.9457, "theoretical_loss": 3.8963958599431874, "tokens_seen": 522518528 }, { "epoch": 0.16, "learning_rate": 0.00042509528585757276, "loss": 2.7995, "theoretical_loss": 3.8963436254198074, "tokens_seen": 522584064 }, { "epoch": 0.16, "objective/train/docs_used": 306243, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.018648862838745, "objective/train/theoretical_loss": 3.8962913992805284, "objective/train/tokens_used": 543109600, "theoretical_loss": 3.8962913992805284, "tokens_seen": 522649600 }, { "epoch": 0.16, "learning_rate": 0.0004250852557673019, "loss": 2.9636, "theoretical_loss": 3.8962913992805284, "tokens_seen": 522649600 }, { "epoch": 0.16, "learning_rate": 0.0004250752256770311, "loss": 2.87, "theoretical_loss": 3.896239181522954, "tokens_seen": 522715136 }, { "epoch": 0.16, "learning_rate": 0.00042506519558676024, "loss": 2.7462, "theoretical_loss": 3.8961869721446885, "tokens_seen": 522780672 }, { "epoch": 0.16, "learning_rate": 0.0004250551654964895, "loss": 2.907, "theoretical_loss": 3.8961347711433376, "tokens_seen": 522846208 }, { "epoch": 0.16, "learning_rate": 0.00042504513540621866, "loss": 2.8195, "theoretical_loss": 3.896082578516506, "tokens_seen": 522911744 }, { "epoch": 0.16, "learning_rate": 0.00042503510531594784, "loss": 2.8101, "theoretical_loss": 3.8960303942618024, "tokens_seen": 522977280 }, { "epoch": 0.16, "learning_rate": 0.000425025075225677, "loss": 3.1211, "theoretical_loss": 3.8959782183768343, "tokens_seen": 523042816 }, { "epoch": 0.16, "learning_rate": 0.0004250150451354062, "loss": 2.6012, "theoretical_loss": 3.8959260508592104, "tokens_seen": 523108352 }, { "epoch": 0.16, "learning_rate": 0.0004250050150451354, "loss": 3.1034, "theoretical_loss": 3.895873891706542, "tokens_seen": 523173888 }, { "epoch": 0.16, "learning_rate": 0.0004249949849548646, "loss": 2.7857, "theoretical_loss": 3.895821740916439, "tokens_seen": 523239424 }, { "epoch": 0.16, "learning_rate": 0.00042498495486459375, "loss": 2.9521, "theoretical_loss": 3.8957695984865137, "tokens_seen": 523304960 }, { "epoch": 0.16, "learning_rate": 0.000424974924774323, "loss": 2.9685, "theoretical_loss": 3.89571746441438, "tokens_seen": 523370496 }, { "epoch": 0.16, "learning_rate": 0.00042496489468405216, "loss": 2.8718, "theoretical_loss": 3.8956653386976514, "tokens_seen": 523436032 }, { "epoch": 0.16, "learning_rate": 0.00042495486459378135, "loss": 2.9434, "theoretical_loss": 3.8956132213339423, "tokens_seen": 523501568 }, { "epoch": 0.16, "learning_rate": 0.0004249448345035105, "loss": 2.9804, "theoretical_loss": 3.8955611123208698, "tokens_seen": 523567104 }, { "epoch": 0.16, "learning_rate": 0.0004249348044132397, "loss": 3.0082, "theoretical_loss": 3.8955090116560505, "tokens_seen": 523632640 }, { "epoch": 0.16, "learning_rate": 0.00042492477432296894, "loss": 3.0848, "theoretical_loss": 3.895456919337102, "tokens_seen": 523698176 }, { "epoch": 0.16, "learning_rate": 0.0004249147442326981, "loss": 2.8986, "theoretical_loss": 3.8954048353616435, "tokens_seen": 523763712 }, { "epoch": 0.16, "learning_rate": 0.0004249047141424273, "loss": 2.6751, "theoretical_loss": 3.8953527597272952, "tokens_seen": 523829248 }, { "epoch": 0.16, "learning_rate": 0.0004248946840521565, "loss": 3.1502, "theoretical_loss": 3.895300692431677, "tokens_seen": 523894784 }, { "epoch": 0.16, "learning_rate": 0.00042488465396188567, "loss": 3.002, "theoretical_loss": 3.8952486334724123, "tokens_seen": 523960320 }, { "epoch": 0.16, "learning_rate": 0.00042487462387161485, "loss": 2.8876, "theoretical_loss": 3.895196582847123, "tokens_seen": 524025856 }, { "epoch": 0.16, "learning_rate": 0.0004248645937813441, "loss": 3.1295, "theoretical_loss": 3.8951445405534324, "tokens_seen": 524091392 }, { "epoch": 0.16, "learning_rate": 0.0004248545636910732, "loss": 3.0447, "theoretical_loss": 3.8950925065889663, "tokens_seen": 524156928 }, { "epoch": 0.16, "learning_rate": 0.00042484453360080245, "loss": 2.6436, "theoretical_loss": 3.8950404809513506, "tokens_seen": 524222464 }, { "epoch": 0.16, "objective/train/docs_used": 306972, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.143721580505371, "objective/train/theoretical_loss": 3.8949884636382106, "objective/train/tokens_used": 544748000, "theoretical_loss": 3.8949884636382106, "tokens_seen": 524288000 }, { "epoch": 0.16, "learning_rate": 0.0004248345035105316, "loss": 2.9599, "theoretical_loss": 3.8949884636382106, "tokens_seen": 524288000 }, { "epoch": 0.16, "learning_rate": 0.0004248244734202608, "loss": 2.937, "theoretical_loss": 3.8949364546471754, "tokens_seen": 524353536 }, { "epoch": 0.16, "learning_rate": 0.00042481444332999, "loss": 2.9483, "theoretical_loss": 3.894884453975873, "tokens_seen": 524419072 }, { "epoch": 0.16, "learning_rate": 0.00042480441323971917, "loss": 2.9053, "theoretical_loss": 3.894832461621934, "tokens_seen": 524484608 }, { "epoch": 0.16, "learning_rate": 0.00042479438314944835, "loss": 2.8771, "theoretical_loss": 3.894780477582988, "tokens_seen": 524550144 }, { "epoch": 0.16, "learning_rate": 0.0004247843530591776, "loss": 2.9409, "theoretical_loss": 3.894728501856667, "tokens_seen": 524615680 }, { "epoch": 0.16, "learning_rate": 0.0004247743229689067, "loss": 3.1486, "theoretical_loss": 3.8946765344406034, "tokens_seen": 524681216 }, { "epoch": 0.16, "learning_rate": 0.00042476429287863595, "loss": 2.5875, "theoretical_loss": 3.8946245753324304, "tokens_seen": 524746752 }, { "epoch": 0.16, "learning_rate": 0.0004247542627883651, "loss": 2.9741, "theoretical_loss": 3.8945726245297836, "tokens_seen": 524812288 }, { "epoch": 0.16, "learning_rate": 0.0004247442326980943, "loss": 2.9664, "theoretical_loss": 3.894520682030297, "tokens_seen": 524877824 }, { "epoch": 0.16, "learning_rate": 0.0004247342026078235, "loss": 2.7809, "theoretical_loss": 3.894468747831609, "tokens_seen": 524943360 }, { "epoch": 0.16, "learning_rate": 0.0004247241725175527, "loss": 3.0679, "theoretical_loss": 3.8944168219313546, "tokens_seen": 525008896 }, { "epoch": 0.16, "learning_rate": 0.00042471414242728186, "loss": 3.0167, "theoretical_loss": 3.8943649043271744, "tokens_seen": 525074432 }, { "epoch": 0.16, "learning_rate": 0.00042470411233701104, "loss": 3.1154, "theoretical_loss": 3.8943129950167066, "tokens_seen": 525139968 }, { "epoch": 0.16, "learning_rate": 0.0004246940822467402, "loss": 3.0939, "theoretical_loss": 3.894261093997592, "tokens_seen": 525205504 }, { "epoch": 0.16, "learning_rate": 0.00042468405215646945, "loss": 3.0354, "theoretical_loss": 3.894209201267471, "tokens_seen": 525271040 }, { "epoch": 0.16, "learning_rate": 0.0004246740220661986, "loss": 3.0519, "theoretical_loss": 3.8941573168239874, "tokens_seen": 525336576 }, { "epoch": 0.16, "learning_rate": 0.0004246639919759278, "loss": 2.9625, "theoretical_loss": 3.8941054406647826, "tokens_seen": 525402112 }, { "epoch": 0.16, "learning_rate": 0.000424653961885657, "loss": 2.8217, "theoretical_loss": 3.894053572787502, "tokens_seen": 525467648 }, { "epoch": 0.16, "learning_rate": 0.0004246439317953862, "loss": 3.094, "theoretical_loss": 3.89400171318979, "tokens_seen": 525533184 }, { "epoch": 0.16, "learning_rate": 0.00042463390170511536, "loss": 2.8674, "theoretical_loss": 3.8939498618692934, "tokens_seen": 525598720 }, { "epoch": 0.16, "learning_rate": 0.00042462387161484454, "loss": 2.8711, "theoretical_loss": 3.893898018823659, "tokens_seen": 525664256 }, { "epoch": 0.16, "learning_rate": 0.0004246138415245737, "loss": 3.068, "theoretical_loss": 3.8938461840505347, "tokens_seen": 525729792 }, { "epoch": 0.16, "learning_rate": 0.00042460381143430296, "loss": 3.1631, "theoretical_loss": 3.8937943575475695, "tokens_seen": 525795328 }, { "epoch": 0.16, "learning_rate": 0.0004245937813440321, "loss": 2.852, "theoretical_loss": 3.893742539312414, "tokens_seen": 525860864 }, { "epoch": 0.16, "objective/train/docs_used": 308263, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0648889541625977, "objective/train/theoretical_loss": 3.8936907293427176, "objective/train/tokens_used": 546386400, "theoretical_loss": 3.8936907293427176, "tokens_seen": 525926400 }, { "epoch": 0.16, "learning_rate": 0.0004245837512537613, "loss": 3.0778, "theoretical_loss": 3.8936907293427176, "tokens_seen": 525926400 }, { "epoch": 0.16, "learning_rate": 0.00042457372116349044, "loss": 2.7427, "theoretical_loss": 3.893638927636134, "tokens_seen": 525991936 }, { "epoch": 0.16, "learning_rate": 0.0004245636910732197, "loss": 3.0738, "theoretical_loss": 3.893587134190314, "tokens_seen": 526057472 }, { "epoch": 0.16, "learning_rate": 0.00042455366098294886, "loss": 2.9898, "theoretical_loss": 3.8935353490029136, "tokens_seen": 526123008 }, { "epoch": 0.16, "learning_rate": 0.00042454363089267804, "loss": 2.9087, "theoretical_loss": 3.893483572071586, "tokens_seen": 526188544 }, { "epoch": 0.16, "learning_rate": 0.0004245336008024072, "loss": 2.8176, "theoretical_loss": 3.8934318033939874, "tokens_seen": 526254080 }, { "epoch": 0.16, "learning_rate": 0.0004245235707121364, "loss": 2.8114, "theoretical_loss": 3.8933800429677743, "tokens_seen": 526319616 }, { "epoch": 0.16, "learning_rate": 0.0004245135406218656, "loss": 2.9196, "theoretical_loss": 3.8933282907906044, "tokens_seen": 526385152 }, { "epoch": 0.16, "learning_rate": 0.0004245035105315948, "loss": 2.8632, "theoretical_loss": 3.8932765468601365, "tokens_seen": 526450688 }, { "epoch": 0.16, "learning_rate": 0.00042449348044132395, "loss": 2.8034, "theoretical_loss": 3.89322481117403, "tokens_seen": 526516224 }, { "epoch": 0.16, "learning_rate": 0.0004244834503510532, "loss": 2.7507, "theoretical_loss": 3.8931730837299456, "tokens_seen": 526581760 }, { "epoch": 0.16, "learning_rate": 0.00042447342026078236, "loss": 2.879, "theoretical_loss": 3.893121364525544, "tokens_seen": 526647296 }, { "epoch": 0.16, "learning_rate": 0.00042446339017051155, "loss": 2.9306, "theoretical_loss": 3.893069653558488, "tokens_seen": 526712832 }, { "epoch": 0.16, "learning_rate": 0.0004244533600802407, "loss": 2.749, "theoretical_loss": 3.893017950826441, "tokens_seen": 526778368 }, { "epoch": 0.16, "learning_rate": 0.0004244433299899699, "loss": 2.88, "theoretical_loss": 3.892966256327068, "tokens_seen": 526843904 }, { "epoch": 0.16, "learning_rate": 0.0004244332998996991, "loss": 3.0049, "theoretical_loss": 3.892914570058033, "tokens_seen": 526909440 }, { "epoch": 0.16, "learning_rate": 0.0004244232698094283, "loss": 2.9474, "theoretical_loss": 3.8928628920170034, "tokens_seen": 526974976 }, { "epoch": 0.16, "learning_rate": 0.00042441323971915745, "loss": 2.9997, "theoretical_loss": 3.8928112222016455, "tokens_seen": 527040512 }, { "epoch": 0.16, "learning_rate": 0.0004244032096288867, "loss": 2.9753, "theoretical_loss": 3.892759560609627, "tokens_seen": 527106048 }, { "epoch": 0.16, "learning_rate": 0.0004243931795386158, "loss": 2.9288, "theoretical_loss": 3.8927079072386186, "tokens_seen": 527171584 }, { "epoch": 0.16, "learning_rate": 0.00042438314944834505, "loss": 2.8508, "theoretical_loss": 3.892656262086289, "tokens_seen": 527237120 }, { "epoch": 0.16, "learning_rate": 0.00042437311935807423, "loss": 2.9228, "theoretical_loss": 3.8926046251503092, "tokens_seen": 527302656 }, { "epoch": 0.16, "learning_rate": 0.0004243630892678034, "loss": 2.9325, "theoretical_loss": 3.8925529964283516, "tokens_seen": 527368192 }, { "epoch": 0.16, "learning_rate": 0.0004243530591775326, "loss": 2.9051, "theoretical_loss": 3.892501375918089, "tokens_seen": 527433728 }, { "epoch": 0.16, "learning_rate": 0.0004243430290872618, "loss": 2.7249, "theoretical_loss": 3.8924497636171953, "tokens_seen": 527499264 }, { "epoch": 0.16, "objective/train/docs_used": 309026, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9705495834350586, "objective/train/theoretical_loss": 3.892398159523345, "objective/train/tokens_used": 548024800, "theoretical_loss": 3.892398159523345, "tokens_seen": 527564800 }, { "epoch": 0.16, "learning_rate": 0.00042433299899699095, "loss": 3.0839, "theoretical_loss": 3.892398159523345, "tokens_seen": 527564800 }, { "epoch": 0.16, "learning_rate": 0.0004243229689067202, "loss": 2.9342, "theoretical_loss": 3.892346563634213, "tokens_seen": 527630336 }, { "epoch": 0.16, "learning_rate": 0.0004243129388164493, "loss": 2.9797, "theoretical_loss": 3.892294975947477, "tokens_seen": 527695872 }, { "epoch": 0.16, "learning_rate": 0.00042430290872617855, "loss": 2.8072, "theoretical_loss": 3.892243396460815, "tokens_seen": 527761408 }, { "epoch": 0.16, "learning_rate": 0.00042429287863590773, "loss": 3.1797, "theoretical_loss": 3.8921918251719054, "tokens_seen": 527826944 }, { "epoch": 0.16, "learning_rate": 0.0004242828485456369, "loss": 2.9326, "theoretical_loss": 3.892140262078427, "tokens_seen": 527892480 }, { "epoch": 0.16, "learning_rate": 0.0004242728184553661, "loss": 2.6238, "theoretical_loss": 3.8920887071780594, "tokens_seen": 527958016 }, { "epoch": 0.16, "learning_rate": 0.0004242627883650953, "loss": 2.8564, "theoretical_loss": 3.8920371604684862, "tokens_seen": 528023552 }, { "epoch": 0.16, "learning_rate": 0.00042425275827482446, "loss": 2.8973, "theoretical_loss": 3.891985621947388, "tokens_seen": 528089088 }, { "epoch": 0.16, "learning_rate": 0.0004242427281845537, "loss": 3.1608, "theoretical_loss": 3.891934091612449, "tokens_seen": 528154624 }, { "epoch": 0.16, "learning_rate": 0.0004242326980942828, "loss": 2.9455, "theoretical_loss": 3.891882569461353, "tokens_seen": 528220160 }, { "epoch": 0.16, "learning_rate": 0.00042422266800401206, "loss": 2.8902, "theoretical_loss": 3.891831055491785, "tokens_seen": 528285696 }, { "epoch": 0.16, "learning_rate": 0.0004242126379137412, "loss": 3.0813, "theoretical_loss": 3.8917795497014316, "tokens_seen": 528351232 }, { "epoch": 0.16, "learning_rate": 0.0004242026078234704, "loss": 2.9741, "theoretical_loss": 3.891728052087979, "tokens_seen": 528416768 }, { "epoch": 0.16, "learning_rate": 0.0004241925777331996, "loss": 2.83, "theoretical_loss": 3.891676562649116, "tokens_seen": 528482304 }, { "epoch": 0.16, "learning_rate": 0.0004241825476429288, "loss": 2.7524, "theoretical_loss": 3.8916250813825313, "tokens_seen": 528547840 }, { "epoch": 0.16, "learning_rate": 0.000424172517552658, "loss": 2.8614, "theoretical_loss": 3.891573608285915, "tokens_seen": 528613376 }, { "epoch": 0.16, "learning_rate": 0.0004241624874623872, "loss": 3.131, "theoretical_loss": 3.891522143356957, "tokens_seen": 528678912 }, { "epoch": 0.16, "learning_rate": 0.0004241524573721164, "loss": 2.6944, "theoretical_loss": 3.89147068659335, "tokens_seen": 528744448 }, { "epoch": 0.16, "learning_rate": 0.00042414242728184556, "loss": 2.948, "theoretical_loss": 3.8914192379927863, "tokens_seen": 528809984 }, { "epoch": 0.16, "learning_rate": 0.00042413239719157474, "loss": 2.9535, "theoretical_loss": 3.891367797552959, "tokens_seen": 528875520 }, { "epoch": 0.16, "learning_rate": 0.0004241223671013039, "loss": 2.9196, "theoretical_loss": 3.891316365271564, "tokens_seen": 528941056 }, { "epoch": 0.16, "learning_rate": 0.00042411233701103316, "loss": 2.8756, "theoretical_loss": 3.891264941146295, "tokens_seen": 529006592 }, { "epoch": 0.16, "learning_rate": 0.0004241023069207623, "loss": 2.8476, "theoretical_loss": 3.89121352517485, "tokens_seen": 529072128 }, { "epoch": 0.16, "learning_rate": 0.0004240922768304915, "loss": 2.9317, "theoretical_loss": 3.8911621173549253, "tokens_seen": 529137664 }, { "epoch": 0.16, "objective/train/docs_used": 310459, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7565929889678955, "objective/train/theoretical_loss": 3.89111071768422, "objective/train/tokens_used": 549663200, "theoretical_loss": 3.89111071768422, "tokens_seen": 529203200 }, { "epoch": 0.16, "learning_rate": 0.00042408224674022065, "loss": 2.9638, "theoretical_loss": 3.89111071768422, "tokens_seen": 529203200 }, { "epoch": 0.16, "learning_rate": 0.0004240722166499499, "loss": 3.0235, "theoretical_loss": 3.8910593261604323, "tokens_seen": 529268736 }, { "epoch": 0.16, "learning_rate": 0.00042406218655967906, "loss": 2.9576, "theoretical_loss": 3.891007942781264, "tokens_seen": 529334272 }, { "epoch": 0.16, "learning_rate": 0.00042405215646940824, "loss": 2.7206, "theoretical_loss": 3.8909565675444138, "tokens_seen": 529399808 }, { "epoch": 0.16, "learning_rate": 0.0004240421263791374, "loss": 2.7056, "theoretical_loss": 3.8909052004475857, "tokens_seen": 529465344 }, { "epoch": 0.16, "learning_rate": 0.0004240320962888666, "loss": 2.6441, "theoretical_loss": 3.8908538414884823, "tokens_seen": 529530880 }, { "epoch": 0.16, "learning_rate": 0.0004240220661985958, "loss": 2.8619, "theoretical_loss": 3.8908024906648073, "tokens_seen": 529596416 }, { "epoch": 0.16, "learning_rate": 0.000424012036108325, "loss": 2.8988, "theoretical_loss": 3.890751147974265, "tokens_seen": 529661952 }, { "epoch": 0.16, "learning_rate": 0.00042400200601805415, "loss": 2.9959, "theoretical_loss": 3.890699813414562, "tokens_seen": 529727488 }, { "epoch": 0.16, "learning_rate": 0.0004239919759277834, "loss": 2.955, "theoretical_loss": 3.890648486983405, "tokens_seen": 529793024 }, { "epoch": 0.16, "learning_rate": 0.00042398194583751256, "loss": 2.841, "theoretical_loss": 3.8905971686785006, "tokens_seen": 529858560 }, { "epoch": 0.16, "learning_rate": 0.00042397191574724175, "loss": 2.9333, "theoretical_loss": 3.8905458584975583, "tokens_seen": 529924096 }, { "epoch": 0.16, "learning_rate": 0.0004239618856569709, "loss": 2.7768, "theoretical_loss": 3.890494556438287, "tokens_seen": 529989632 }, { "epoch": 0.16, "learning_rate": 0.0004239518555667001, "loss": 2.913, "theoretical_loss": 3.8904432624983976, "tokens_seen": 530055168 }, { "epoch": 0.16, "learning_rate": 0.0004239418254764293, "loss": 2.9885, "theoretical_loss": 3.890391976675601, "tokens_seen": 530120704 }, { "epoch": 0.16, "learning_rate": 0.0004239317953861585, "loss": 2.9349, "theoretical_loss": 3.89034069896761, "tokens_seen": 530186240 }, { "epoch": 0.16, "learning_rate": 0.00042392176529588765, "loss": 2.5791, "theoretical_loss": 3.890289429372137, "tokens_seen": 530251776 }, { "epoch": 0.16, "learning_rate": 0.0004239117352056169, "loss": 3.0469, "theoretical_loss": 3.8902381678868974, "tokens_seen": 530317312 }, { "epoch": 0.16, "learning_rate": 0.000423901705115346, "loss": 2.8172, "theoretical_loss": 3.890186914509605, "tokens_seen": 530382848 }, { "epoch": 0.16, "learning_rate": 0.00042389167502507525, "loss": 2.8591, "theoretical_loss": 3.890135669237976, "tokens_seen": 530448384 }, { "epoch": 0.16, "learning_rate": 0.00042388164493480443, "loss": 3.1571, "theoretical_loss": 3.8900844320697274, "tokens_seen": 530513920 }, { "epoch": 0.16, "learning_rate": 0.0004238716148445336, "loss": 3.1662, "theoretical_loss": 3.8900332030025773, "tokens_seen": 530579456 }, { "epoch": 0.16, "learning_rate": 0.0004238615847542628, "loss": 2.862, "theoretical_loss": 3.8899819820342443, "tokens_seen": 530644992 }, { "epoch": 0.16, "learning_rate": 0.000423851554663992, "loss": 2.7471, "theoretical_loss": 3.889930769162448, "tokens_seen": 530710528 }, { "epoch": 0.16, "learning_rate": 0.00042384152457372115, "loss": 2.9666, "theoretical_loss": 3.8898795643849087, "tokens_seen": 530776064 }, { "epoch": 0.16, "objective/train/docs_used": 311126, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.5528128147125244, "objective/train/theoretical_loss": 3.889828367699349, "objective/train/tokens_used": 551301600, "theoretical_loss": 3.889828367699349, "tokens_seen": 530841600 }, { "epoch": 0.16, "learning_rate": 0.0004238314944834504, "loss": 3.1406, "theoretical_loss": 3.889828367699349, "tokens_seen": 530841600 }, { "epoch": 0.16, "learning_rate": 0.0004238214643931795, "loss": 2.7622, "theoretical_loss": 3.88977717910349, "tokens_seen": 530907136 }, { "epoch": 0.16, "learning_rate": 0.00042381143430290875, "loss": 2.8677, "theoretical_loss": 3.8897259985950554, "tokens_seen": 530972672 }, { "epoch": 0.16, "learning_rate": 0.00042380140421263793, "loss": 2.9447, "theoretical_loss": 3.8896748261717695, "tokens_seen": 531038208 }, { "epoch": 0.16, "learning_rate": 0.0004237913741223671, "loss": 2.9975, "theoretical_loss": 3.889623661831359, "tokens_seen": 531103744 }, { "epoch": 0.16, "learning_rate": 0.0004237813440320963, "loss": 2.8501, "theoretical_loss": 3.8895725055715475, "tokens_seen": 531169280 }, { "epoch": 0.16, "learning_rate": 0.0004237713139418255, "loss": 2.8257, "theoretical_loss": 3.8895213573900635, "tokens_seen": 531234816 }, { "epoch": 0.16, "learning_rate": 0.00042376128385155466, "loss": 2.8109, "theoretical_loss": 3.889470217284635, "tokens_seen": 531300352 }, { "epoch": 0.16, "learning_rate": 0.0004237512537612839, "loss": 2.7859, "theoretical_loss": 3.8894190852529906, "tokens_seen": 531365888 }, { "epoch": 0.16, "learning_rate": 0.000423741223671013, "loss": 2.8411, "theoretical_loss": 3.8893679612928604, "tokens_seen": 531431424 }, { "epoch": 0.16, "learning_rate": 0.00042373119358074226, "loss": 2.7203, "theoretical_loss": 3.889316845401974, "tokens_seen": 531496960 }, { "epoch": 0.16, "learning_rate": 0.0004237211634904714, "loss": 2.5823, "theoretical_loss": 3.8892657375780644, "tokens_seen": 531562496 }, { "epoch": 0.16, "learning_rate": 0.0004237111334002006, "loss": 2.8587, "theoretical_loss": 3.8892146378188643, "tokens_seen": 531628032 }, { "epoch": 0.16, "learning_rate": 0.0004237011033099298, "loss": 2.8946, "theoretical_loss": 3.8891635461221057, "tokens_seen": 531693568 }, { "epoch": 0.16, "learning_rate": 0.000423691073219659, "loss": 2.8974, "theoretical_loss": 3.8891124624855236, "tokens_seen": 531759104 }, { "epoch": 0.16, "learning_rate": 0.00042368104312938816, "loss": 2.7539, "theoretical_loss": 3.889061386906854, "tokens_seen": 531824640 }, { "epoch": 0.16, "learning_rate": 0.0004236710130391174, "loss": 2.7946, "theoretical_loss": 3.889010319383832, "tokens_seen": 531890176 }, { "epoch": 0.16, "learning_rate": 0.0004236609829488465, "loss": 3.0662, "theoretical_loss": 3.8889592599141958, "tokens_seen": 531955712 }, { "epoch": 0.16, "learning_rate": 0.00042365095285857576, "loss": 2.9279, "theoretical_loss": 3.888908208495683, "tokens_seen": 532021248 }, { "epoch": 0.16, "learning_rate": 0.0004236409227683049, "loss": 2.8842, "theoretical_loss": 3.8888571651260326, "tokens_seen": 532086784 }, { "epoch": 0.16, "learning_rate": 0.0004236308926780341, "loss": 2.6855, "theoretical_loss": 3.888806129802984, "tokens_seen": 532152320 }, { "epoch": 0.16, "learning_rate": 0.0004236208625877633, "loss": 2.9097, "theoretical_loss": 3.888755102524279, "tokens_seen": 532217856 }, { "epoch": 0.16, "learning_rate": 0.0004236108324974925, "loss": 2.718, "theoretical_loss": 3.8887040832876587, "tokens_seen": 532283392 }, { "epoch": 0.16, "learning_rate": 0.00042360080240722166, "loss": 3.062, "theoretical_loss": 3.8886530720908654, "tokens_seen": 532348928 }, { "epoch": 0.16, "learning_rate": 0.00042359077231695085, "loss": 2.9053, "theoretical_loss": 3.888602068931643, "tokens_seen": 532414464 }, { "epoch": 0.16, "objective/train/docs_used": 312276, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.722914934158325, "objective/train/theoretical_loss": 3.888551073807736, "objective/train/tokens_used": 552940000, "theoretical_loss": 3.888551073807736, "tokens_seen": 532480000 }, { "epoch": 0.16, "learning_rate": 0.00042358074222668, "loss": 2.8273, "theoretical_loss": 3.888551073807736, "tokens_seen": 532480000 }, { "epoch": 0.16, "learning_rate": 0.00042357071213640926, "loss": 2.8933, "theoretical_loss": 3.8885000867168893, "tokens_seen": 532545536 }, { "epoch": 0.16, "learning_rate": 0.0004235606820461384, "loss": 2.7793, "theoretical_loss": 3.88844910765685, "tokens_seen": 532611072 }, { "epoch": 0.16, "learning_rate": 0.0004235506519558676, "loss": 2.9064, "theoretical_loss": 3.888398136625364, "tokens_seen": 532676608 }, { "epoch": 0.16, "learning_rate": 0.00042354062186559675, "loss": 3.0965, "theoretical_loss": 3.888347173620181, "tokens_seen": 532742144 }, { "epoch": 0.16, "learning_rate": 0.000423530591775326, "loss": 2.8523, "theoretical_loss": 3.8882962186390486, "tokens_seen": 532807680 }, { "epoch": 0.16, "learning_rate": 0.00042352056168505517, "loss": 3.0964, "theoretical_loss": 3.8882452716797173, "tokens_seen": 532873216 }, { "epoch": 0.16, "learning_rate": 0.00042351053159478435, "loss": 2.7862, "theoretical_loss": 3.8881943327399378, "tokens_seen": 532938752 }, { "epoch": 0.16, "learning_rate": 0.00042350050150451353, "loss": 3.0715, "theoretical_loss": 3.8881434018174614, "tokens_seen": 533004288 }, { "epoch": 0.16, "learning_rate": 0.00042349047141424276, "loss": 3.2114, "theoretical_loss": 3.8880924789100417, "tokens_seen": 533069824 }, { "epoch": 0.16, "learning_rate": 0.0004234804413239719, "loss": 3.0219, "theoretical_loss": 3.888041564015431, "tokens_seen": 533135360 }, { "epoch": 0.16, "learning_rate": 0.00042347041123370113, "loss": 2.7477, "theoretical_loss": 3.8879906571313847, "tokens_seen": 533200896 }, { "epoch": 0.16, "learning_rate": 0.00042346038114343025, "loss": 2.921, "theoretical_loss": 3.887939758255658, "tokens_seen": 533266432 }, { "epoch": 0.16, "learning_rate": 0.0004234503510531595, "loss": 2.9896, "theoretical_loss": 3.887888867386007, "tokens_seen": 533331968 }, { "epoch": 0.16, "learning_rate": 0.00042344032096288867, "loss": 2.789, "theoretical_loss": 3.8878379845201883, "tokens_seen": 533397504 }, { "epoch": 0.16, "learning_rate": 0.00042343029087261785, "loss": 2.8581, "theoretical_loss": 3.88778710965596, "tokens_seen": 533463040 }, { "epoch": 0.16, "learning_rate": 0.0004234202607823471, "loss": 3.0571, "theoretical_loss": 3.887736242791082, "tokens_seen": 533528576 }, { "epoch": 0.16, "learning_rate": 0.0004234102306920762, "loss": 2.8806, "theoretical_loss": 3.887685383923314, "tokens_seen": 533594112 }, { "epoch": 0.16, "learning_rate": 0.00042340020060180545, "loss": 2.9129, "theoretical_loss": 3.8876345330504156, "tokens_seen": 533659648 }, { "epoch": 0.16, "learning_rate": 0.00042339017051153463, "loss": 3.0494, "theoretical_loss": 3.887583690170149, "tokens_seen": 533725184 }, { "epoch": 0.16, "learning_rate": 0.0004233801404212638, "loss": 2.7722, "theoretical_loss": 3.887532855280277, "tokens_seen": 533790720 }, { "epoch": 0.16, "learning_rate": 0.000423370110330993, "loss": 2.6642, "theoretical_loss": 3.887482028378564, "tokens_seen": 533856256 }, { "epoch": 0.16, "learning_rate": 0.0004233600802407222, "loss": 2.8821, "theoretical_loss": 3.8874312094627723, "tokens_seen": 533921792 }, { "epoch": 0.16, "learning_rate": 0.00042335005015045135, "loss": 2.8685, "theoretical_loss": 3.8873803985306683, "tokens_seen": 533987328 }, { "epoch": 0.16, "learning_rate": 0.0004233400200601806, "loss": 2.7004, "theoretical_loss": 3.8873295955800184, "tokens_seen": 534052864 }, { "epoch": 0.16, "objective/train/docs_used": 312792, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.320737838745117, "objective/train/theoretical_loss": 3.8872788006085894, "objective/train/tokens_used": 554578400, "theoretical_loss": 3.8872788006085894, "tokens_seen": 534118400 }, { "epoch": 0.16, "learning_rate": 0.0004233299899699097, "loss": 2.8152, "theoretical_loss": 3.8872788006085894, "tokens_seen": 534118400 }, { "epoch": 0.16, "learning_rate": 0.00042331995987963895, "loss": 3.1119, "theoretical_loss": 3.8872280136141484, "tokens_seen": 534183936 }, { "epoch": 0.16, "learning_rate": 0.00042330992978936813, "loss": 2.8227, "theoretical_loss": 3.887177234594466, "tokens_seen": 534249472 }, { "epoch": 0.16, "learning_rate": 0.0004232998996990973, "loss": 2.7367, "theoretical_loss": 3.88712646354731, "tokens_seen": 534315008 }, { "epoch": 0.16, "learning_rate": 0.0004232898696088265, "loss": 2.8633, "theoretical_loss": 3.8870757004704526, "tokens_seen": 534380544 }, { "epoch": 0.16, "learning_rate": 0.0004232798395185557, "loss": 2.9422, "theoretical_loss": 3.8870249453616648, "tokens_seen": 534446080 }, { "epoch": 0.16, "learning_rate": 0.00042326980942828486, "loss": 2.8519, "theoretical_loss": 3.8869741982187183, "tokens_seen": 534511616 }, { "epoch": 0.16, "learning_rate": 0.0004232597793380141, "loss": 3.0843, "theoretical_loss": 3.8869234590393873, "tokens_seen": 534577152 }, { "epoch": 0.16, "learning_rate": 0.0004232497492477432, "loss": 3.0001, "theoretical_loss": 3.886872727821446, "tokens_seen": 534642688 }, { "epoch": 0.16, "learning_rate": 0.00042323971915747246, "loss": 2.8419, "theoretical_loss": 3.88682200456267, "tokens_seen": 534708224 }, { "epoch": 0.16, "learning_rate": 0.0004232296890672016, "loss": 2.9222, "theoretical_loss": 3.886771289260834, "tokens_seen": 534773760 }, { "epoch": 0.16, "learning_rate": 0.0004232196589769308, "loss": 2.9703, "theoretical_loss": 3.886720581913716, "tokens_seen": 534839296 }, { "epoch": 0.16, "learning_rate": 0.00042320962888666, "loss": 3.1677, "theoretical_loss": 3.886669882519093, "tokens_seen": 534904832 }, { "epoch": 0.16, "learning_rate": 0.0004231995987963892, "loss": 2.9986, "theoretical_loss": 3.886619191074744, "tokens_seen": 534970368 }, { "epoch": 0.16, "learning_rate": 0.00042318956870611836, "loss": 2.9794, "theoretical_loss": 3.886568507578448, "tokens_seen": 535035904 }, { "epoch": 0.16, "learning_rate": 0.0004231795386158476, "loss": 2.9556, "theoretical_loss": 3.8865178320279874, "tokens_seen": 535101440 }, { "epoch": 0.16, "learning_rate": 0.0004231695085255767, "loss": 3.1085, "theoretical_loss": 3.886467164421142, "tokens_seen": 535166976 }, { "epoch": 0.16, "learning_rate": 0.00042315947843530596, "loss": 2.8245, "theoretical_loss": 3.8864165047556942, "tokens_seen": 535232512 }, { "epoch": 0.16, "learning_rate": 0.0004231494483450351, "loss": 2.9032, "theoretical_loss": 3.886365853029427, "tokens_seen": 535298048 }, { "epoch": 0.16, "learning_rate": 0.0004231394182547643, "loss": 2.9161, "theoretical_loss": 3.8863152092401254, "tokens_seen": 535363584 }, { "epoch": 0.16, "learning_rate": 0.0004231293881644935, "loss": 2.9488, "theoretical_loss": 3.886264573385574, "tokens_seen": 535429120 }, { "epoch": 0.16, "learning_rate": 0.0004231193580742227, "loss": 2.9453, "theoretical_loss": 3.886213945463558, "tokens_seen": 535494656 }, { "epoch": 0.16, "learning_rate": 0.00042310932798395186, "loss": 3.054, "theoretical_loss": 3.886163325471864, "tokens_seen": 535560192 }, { "epoch": 0.16, "learning_rate": 0.00042309929789368105, "loss": 2.8614, "theoretical_loss": 3.8861127134082807, "tokens_seen": 535625728 }, { "epoch": 0.16, "learning_rate": 0.0004230892678034102, "loss": 3.0729, "theoretical_loss": 3.8860621092705956, "tokens_seen": 535691264 }, { "epoch": 0.16, "objective/train/docs_used": 314102, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.790719747543335, "objective/train/theoretical_loss": 3.8860115130565984, "objective/train/tokens_used": 556216800, "theoretical_loss": 3.8860115130565984, "tokens_seen": 535756800 }, { "epoch": 0.16, "learning_rate": 0.00042307923771313946, "loss": 3.0928, "theoretical_loss": 3.8860115130565984, "tokens_seen": 535756800 }, { "epoch": 0.16, "learning_rate": 0.0004230692076228686, "loss": 2.9227, "theoretical_loss": 3.8859609247640794, "tokens_seen": 535822336 }, { "epoch": 0.16, "learning_rate": 0.0004230591775325978, "loss": 2.7427, "theoretical_loss": 3.88591034439083, "tokens_seen": 535887872 }, { "epoch": 0.16, "learning_rate": 0.00042304914744232695, "loss": 3.0429, "theoretical_loss": 3.885859771934642, "tokens_seen": 535953408 }, { "epoch": 0.16, "learning_rate": 0.0004230391173520562, "loss": 2.8292, "theoretical_loss": 3.885809207393308, "tokens_seen": 536018944 }, { "epoch": 0.16, "learning_rate": 0.00042302908726178537, "loss": 2.798, "theoretical_loss": 3.8857586507646227, "tokens_seen": 536084480 }, { "epoch": 0.16, "learning_rate": 0.00042301905717151455, "loss": 2.9397, "theoretical_loss": 3.88570810204638, "tokens_seen": 536150016 }, { "epoch": 0.16, "learning_rate": 0.00042300902708124373, "loss": 2.7327, "theoretical_loss": 3.8856575612363753, "tokens_seen": 536215552 }, { "epoch": 0.16, "learning_rate": 0.00042299899699097297, "loss": 2.9514, "theoretical_loss": 3.8856070283324056, "tokens_seen": 536281088 }, { "epoch": 0.16, "learning_rate": 0.0004229889669007021, "loss": 2.7247, "theoretical_loss": 3.885556503332268, "tokens_seen": 536346624 }, { "epoch": 0.16, "learning_rate": 0.00042297893681043133, "loss": 3.0424, "theoretical_loss": 3.8855059862337606, "tokens_seen": 536412160 }, { "epoch": 0.16, "learning_rate": 0.00042296890672016045, "loss": 3.0064, "theoretical_loss": 3.885455477034683, "tokens_seen": 536477696 }, { "epoch": 0.16, "learning_rate": 0.0004229588766298897, "loss": 2.8565, "theoretical_loss": 3.8854049757328353, "tokens_seen": 536543232 }, { "epoch": 0.16, "learning_rate": 0.00042294884653961887, "loss": 2.9559, "theoretical_loss": 3.8853544823260178, "tokens_seen": 536608768 }, { "epoch": 0.16, "learning_rate": 0.00042293881644934805, "loss": 2.8287, "theoretical_loss": 3.8853039968120324, "tokens_seen": 536674304 }, { "epoch": 0.16, "learning_rate": 0.00042292878635907723, "loss": 2.909, "theoretical_loss": 3.885253519188682, "tokens_seen": 536739840 }, { "epoch": 0.16, "learning_rate": 0.0004229187562688064, "loss": 3.0058, "theoretical_loss": 3.885203049453769, "tokens_seen": 536805376 }, { "epoch": 0.16, "learning_rate": 0.0004229087261785356, "loss": 2.9078, "theoretical_loss": 3.8851525876050994, "tokens_seen": 536870912 }, { "epoch": 0.16, "learning_rate": 0.00042289869608826483, "loss": 2.6801, "theoretical_loss": 3.885102133640477, "tokens_seen": 536936448 }, { "epoch": 0.16, "learning_rate": 0.00042288866599799396, "loss": 2.8486, "theoretical_loss": 3.8850516875577092, "tokens_seen": 537001984 }, { "epoch": 0.16, "learning_rate": 0.0004228786359077232, "loss": 2.9838, "theoretical_loss": 3.8850012493546027, "tokens_seen": 537067520 }, { "epoch": 0.16, "learning_rate": 0.0004228686058174523, "loss": 2.9971, "theoretical_loss": 3.884950819028965, "tokens_seen": 537133056 }, { "epoch": 0.16, "learning_rate": 0.00042285857572718155, "loss": 2.7513, "theoretical_loss": 3.884900396578605, "tokens_seen": 537198592 }, { "epoch": 0.16, "learning_rate": 0.00042284854563691074, "loss": 2.9549, "theoretical_loss": 3.8848499820013327, "tokens_seen": 537264128 }, { "epoch": 0.16, "learning_rate": 0.0004228385155466399, "loss": 2.9562, "theoretical_loss": 3.884799575294958, "tokens_seen": 537329664 }, { "epoch": 0.16, "objective/train/docs_used": 314596, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.16971755027771, "objective/train/theoretical_loss": 3.8847491764572926, "objective/train/tokens_used": 557855200, "theoretical_loss": 3.8847491764572926, "tokens_seen": 537395200 }, { "epoch": 0.16, "learning_rate": 0.0004228284854563691, "loss": 2.9893, "theoretical_loss": 3.8847491764572926, "tokens_seen": 537395200 }, { "epoch": 0.16, "learning_rate": 0.00042281845536609833, "loss": 2.9383, "theoretical_loss": 3.884698785486149, "tokens_seen": 537460736 }, { "epoch": 0.16, "learning_rate": 0.00042280842527582746, "loss": 2.5575, "theoretical_loss": 3.884648402379341, "tokens_seen": 537526272 }, { "epoch": 0.16, "learning_rate": 0.0004227983951855567, "loss": 2.6068, "theoretical_loss": 3.8845980271346816, "tokens_seen": 537591808 }, { "epoch": 0.16, "learning_rate": 0.0004227883650952858, "loss": 2.8686, "theoretical_loss": 3.8845476597499857, "tokens_seen": 537657344 }, { "epoch": 0.16, "learning_rate": 0.00042277833500501506, "loss": 3.0102, "theoretical_loss": 3.884497300223069, "tokens_seen": 537722880 }, { "epoch": 0.16, "learning_rate": 0.00042276830491474424, "loss": 2.943, "theoretical_loss": 3.8844469485517488, "tokens_seen": 537788416 }, { "epoch": 0.16, "learning_rate": 0.0004227582748244734, "loss": 2.748, "theoretical_loss": 3.8843966047338423, "tokens_seen": 537853952 }, { "epoch": 0.16, "learning_rate": 0.0004227482447342026, "loss": 3.0621, "theoretical_loss": 3.8843462687671684, "tokens_seen": 537919488 }, { "epoch": 0.16, "learning_rate": 0.0004227382146439318, "loss": 2.7502, "theoretical_loss": 3.8842959406495456, "tokens_seen": 537985024 }, { "epoch": 0.16, "learning_rate": 0.00042272818455366096, "loss": 3.0579, "theoretical_loss": 3.884245620378794, "tokens_seen": 538050560 }, { "epoch": 0.16, "learning_rate": 0.0004227181544633902, "loss": 2.8916, "theoretical_loss": 3.8841953079527354, "tokens_seen": 538116096 }, { "epoch": 0.16, "learning_rate": 0.0004227081243731193, "loss": 2.8935, "theoretical_loss": 3.8841450033691913, "tokens_seen": 538181632 }, { "epoch": 0.16, "learning_rate": 0.00042269809428284856, "loss": 2.9873, "theoretical_loss": 3.8840947066259845, "tokens_seen": 538247168 }, { "epoch": 0.16, "learning_rate": 0.0004226880641925777, "loss": 2.7005, "theoretical_loss": 3.884044417720938, "tokens_seen": 538312704 }, { "epoch": 0.16, "learning_rate": 0.0004226780341023069, "loss": 2.5778, "theoretical_loss": 3.8839941366518778, "tokens_seen": 538378240 }, { "epoch": 0.16, "learning_rate": 0.00042266800401203616, "loss": 2.8099, "theoretical_loss": 3.8839438634166275, "tokens_seen": 538443776 }, { "epoch": 0.16, "learning_rate": 0.0004226579739217653, "loss": 2.8237, "theoretical_loss": 3.883893598013014, "tokens_seen": 538509312 }, { "epoch": 0.16, "learning_rate": 0.0004226479438314945, "loss": 2.7426, "theoretical_loss": 3.883843340438865, "tokens_seen": 538574848 }, { "epoch": 0.16, "learning_rate": 0.0004226379137412237, "loss": 2.84, "theoretical_loss": 3.8837930906920075, "tokens_seen": 538640384 }, { "epoch": 0.16, "learning_rate": 0.0004226278836509529, "loss": 2.5758, "theoretical_loss": 3.8837428487702716, "tokens_seen": 538705920 }, { "epoch": 0.16, "learning_rate": 0.00042261785356068206, "loss": 2.9282, "theoretical_loss": 3.8836926146714856, "tokens_seen": 538771456 }, { "epoch": 0.16, "learning_rate": 0.00042260782347041125, "loss": 2.6134, "theoretical_loss": 3.883642388393481, "tokens_seen": 538836992 }, { "epoch": 0.16, "learning_rate": 0.0004225977933801404, "loss": 2.821, "theoretical_loss": 3.8835921699340887, "tokens_seen": 538902528 }, { "epoch": 0.16, "learning_rate": 0.00042258776328986966, "loss": 2.9375, "theoretical_loss": 3.8835419592911413, "tokens_seen": 538968064 }, { "epoch": 0.16, "objective/train/docs_used": 315427, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6693382263183594, "objective/train/theoretical_loss": 3.8834917564624716, "objective/train/tokens_used": 559493600, "theoretical_loss": 3.8834917564624716, "tokens_seen": 539033600 }, { "epoch": 0.16, "learning_rate": 0.0004225777331995988, "loss": 2.8935, "theoretical_loss": 3.8834917564624716, "tokens_seen": 539033600 }, { "epoch": 0.16, "learning_rate": 0.000422567703109328, "loss": 2.8414, "theoretical_loss": 3.883441561445914, "tokens_seen": 539099136 }, { "epoch": 0.16, "learning_rate": 0.00042255767301905715, "loss": 3.1531, "theoretical_loss": 3.8833913742393036, "tokens_seen": 539164672 }, { "epoch": 0.16, "learning_rate": 0.0004225476429287864, "loss": 2.7758, "theoretical_loss": 3.8833411948404755, "tokens_seen": 539230208 }, { "epoch": 0.16, "learning_rate": 0.00042253761283851557, "loss": 2.9681, "theoretical_loss": 3.883291023247267, "tokens_seen": 539295744 }, { "epoch": 0.16, "learning_rate": 0.00042252758274824475, "loss": 2.8058, "theoretical_loss": 3.8832408594575147, "tokens_seen": 539361280 }, { "epoch": 0.16, "learning_rate": 0.00042251755265797393, "loss": 2.8653, "theoretical_loss": 3.883190703469058, "tokens_seen": 539426816 }, { "epoch": 0.16, "learning_rate": 0.00042250752256770317, "loss": 2.8425, "theoretical_loss": 3.883140555279735, "tokens_seen": 539492352 }, { "epoch": 0.16, "learning_rate": 0.0004224974924774323, "loss": 2.942, "theoretical_loss": 3.883090414887387, "tokens_seen": 539557888 }, { "epoch": 0.16, "learning_rate": 0.00042248746238716153, "loss": 3.0772, "theoretical_loss": 3.8830402822898535, "tokens_seen": 539623424 }, { "epoch": 0.16, "learning_rate": 0.00042247743229689065, "loss": 2.6957, "theoretical_loss": 3.8829901574849774, "tokens_seen": 539688960 }, { "epoch": 0.16, "learning_rate": 0.0004224674022066199, "loss": 2.8871, "theoretical_loss": 3.8829400404706003, "tokens_seen": 539754496 }, { "epoch": 0.16, "learning_rate": 0.00042245737211634907, "loss": 2.855, "theoretical_loss": 3.8828899312445673, "tokens_seen": 539820032 }, { "epoch": 0.16, "learning_rate": 0.00042244734202607825, "loss": 2.9288, "theoretical_loss": 3.8828398298047206, "tokens_seen": 539885568 }, { "epoch": 0.16, "learning_rate": 0.00042243731193580743, "loss": 3.0415, "theoretical_loss": 3.8827897361489074, "tokens_seen": 539951104 }, { "epoch": 0.16, "learning_rate": 0.0004224272818455366, "loss": 2.9481, "theoretical_loss": 3.8827396502749734, "tokens_seen": 540016640 }, { "epoch": 0.16, "learning_rate": 0.0004224172517552658, "loss": 3.0045, "theoretical_loss": 3.882689572180764, "tokens_seen": 540082176 }, { "epoch": 0.16, "learning_rate": 0.00042240722166499503, "loss": 2.6324, "theoretical_loss": 3.882639501864129, "tokens_seen": 540147712 }, { "epoch": 0.16, "learning_rate": 0.00042239719157472416, "loss": 2.7887, "theoretical_loss": 3.8825894393229152, "tokens_seen": 540213248 }, { "epoch": 0.16, "learning_rate": 0.0004223871614844534, "loss": 2.7621, "theoretical_loss": 3.882539384554974, "tokens_seen": 540278784 }, { "epoch": 0.16, "learning_rate": 0.0004223771313941825, "loss": 2.8169, "theoretical_loss": 3.8824893375581544, "tokens_seen": 540344320 }, { "epoch": 0.16, "learning_rate": 0.00042236710130391176, "loss": 2.771, "theoretical_loss": 3.8824392983303078, "tokens_seen": 540409856 }, { "epoch": 0.16, "learning_rate": 0.00042235707121364094, "loss": 2.7474, "theoretical_loss": 3.8823892668692865, "tokens_seen": 540475392 }, { "epoch": 0.16, "learning_rate": 0.0004223470411233701, "loss": 3.0539, "theoretical_loss": 3.8823392431729435, "tokens_seen": 540540928 }, { "epoch": 0.16, "learning_rate": 0.0004223370110330993, "loss": 2.9556, "theoretical_loss": 3.8822892272391325, "tokens_seen": 540606464 }, { "epoch": 0.16, "objective/train/docs_used": 316530, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7001001834869385, "objective/train/theoretical_loss": 3.882239219065708, "objective/train/tokens_used": 561132000, "theoretical_loss": 3.882239219065708, "tokens_seen": 540672000 }, { "epoch": 0.16, "learning_rate": 0.00042232698094282853, "loss": 2.8921, "theoretical_loss": 3.882239219065708, "tokens_seen": 540672000 }, { "epoch": 0.16, "learning_rate": 0.00042231695085255766, "loss": 2.7654, "theoretical_loss": 3.882189218650525, "tokens_seen": 540737536 }, { "epoch": 0.16, "learning_rate": 0.0004223069207622869, "loss": 2.7423, "theoretical_loss": 3.882139225991441, "tokens_seen": 540803072 }, { "epoch": 0.16, "learning_rate": 0.000422296890672016, "loss": 2.752, "theoretical_loss": 3.8820892410863124, "tokens_seen": 540868608 }, { "epoch": 0.16, "learning_rate": 0.00042228686058174526, "loss": 2.8386, "theoretical_loss": 3.882039263932997, "tokens_seen": 540934144 }, { "epoch": 0.16, "learning_rate": 0.00042227683049147444, "loss": 2.7466, "theoretical_loss": 3.8819892945293546, "tokens_seen": 540999680 }, { "epoch": 0.16, "learning_rate": 0.0004222668004012036, "loss": 3.0137, "theoretical_loss": 3.8819393328732446, "tokens_seen": 541065216 }, { "epoch": 0.16, "learning_rate": 0.0004222567703109328, "loss": 2.9363, "theoretical_loss": 3.8818893789625273, "tokens_seen": 541130752 }, { "epoch": 0.16, "learning_rate": 0.000422246740220662, "loss": 2.821, "theoretical_loss": 3.881839432795064, "tokens_seen": 541196288 }, { "epoch": 0.16, "learning_rate": 0.00042223671013039116, "loss": 2.8432, "theoretical_loss": 3.881789494368717, "tokens_seen": 541261824 }, { "epoch": 0.16, "learning_rate": 0.0004222266800401204, "loss": 2.7617, "theoretical_loss": 3.88173956368135, "tokens_seen": 541327360 }, { "epoch": 0.16, "learning_rate": 0.0004222166499498495, "loss": 2.9612, "theoretical_loss": 3.8816896407308263, "tokens_seen": 541392896 }, { "epoch": 0.16, "learning_rate": 0.00042220661985957876, "loss": 2.7397, "theoretical_loss": 3.8816397255150115, "tokens_seen": 541458432 }, { "epoch": 0.16, "learning_rate": 0.0004221965897693079, "loss": 2.9595, "theoretical_loss": 3.8815898180317703, "tokens_seen": 541523968 }, { "epoch": 0.16, "learning_rate": 0.0004221865596790371, "loss": 2.9104, "theoretical_loss": 3.8815399182789703, "tokens_seen": 541589504 }, { "epoch": 0.16, "learning_rate": 0.0004221765295887663, "loss": 2.7983, "theoretical_loss": 3.881490026254478, "tokens_seen": 541655040 }, { "epoch": 0.16, "learning_rate": 0.0004221664994984955, "loss": 2.832, "theoretical_loss": 3.881440141956162, "tokens_seen": 541720576 }, { "epoch": 0.16, "learning_rate": 0.00042215646940822467, "loss": 2.8153, "theoretical_loss": 3.8813902653818912, "tokens_seen": 541786112 }, { "epoch": 0.16, "learning_rate": 0.0004221464393179539, "loss": 2.8717, "theoretical_loss": 3.8813403965295357, "tokens_seen": 541851648 }, { "epoch": 0.16, "learning_rate": 0.00042213640922768303, "loss": 2.8259, "theoretical_loss": 3.8812905353969667, "tokens_seen": 541917184 }, { "epoch": 0.16, "learning_rate": 0.00042212637913741226, "loss": 2.8444, "theoretical_loss": 3.881240681982054, "tokens_seen": 541982720 }, { "epoch": 0.16, "learning_rate": 0.0004221163490471414, "loss": 2.8232, "theoretical_loss": 3.8811908362826726, "tokens_seen": 542048256 }, { "epoch": 0.16, "learning_rate": 0.0004221063189568706, "loss": 2.9688, "theoretical_loss": 3.881140998296694, "tokens_seen": 542113792 }, { "epoch": 0.16, "learning_rate": 0.0004220962888665998, "loss": 2.7413, "theoretical_loss": 3.8810911680219933, "tokens_seen": 542179328 }, { "epoch": 0.16, "learning_rate": 0.000422086258776329, "loss": 2.762, "theoretical_loss": 3.881041345456444, "tokens_seen": 542244864 }, { "epoch": 0.16, "objective/train/docs_used": 317270, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.066617727279663, "objective/train/theoretical_loss": 3.8809915305979237, "objective/train/tokens_used": 562770400, "theoretical_loss": 3.8809915305979237, "tokens_seen": 542310400 }, { "epoch": 0.16, "learning_rate": 0.00042207622868605817, "loss": 2.8901, "theoretical_loss": 3.8809915305979237, "tokens_seen": 542310400 }, { "epoch": 0.16, "learning_rate": 0.00042206619859578735, "loss": 2.9818, "theoretical_loss": 3.880941723444308, "tokens_seen": 542375936 }, { "epoch": 0.16, "learning_rate": 0.00042205616850551653, "loss": 2.7985, "theoretical_loss": 3.8808919239934747, "tokens_seen": 542441472 }, { "epoch": 0.16, "learning_rate": 0.00042204613841524577, "loss": 2.7908, "theoretical_loss": 3.8808421322433024, "tokens_seen": 542507008 }, { "epoch": 0.16, "learning_rate": 0.0004220361083249749, "loss": 2.7691, "theoretical_loss": 3.8807923481916697, "tokens_seen": 542572544 }, { "epoch": 0.16, "learning_rate": 0.00042202607823470413, "loss": 2.7955, "theoretical_loss": 3.880742571836457, "tokens_seen": 542638080 }, { "epoch": 0.16, "learning_rate": 0.0004220160481444333, "loss": 2.6381, "theoretical_loss": 3.8806928031755454, "tokens_seen": 542703616 }, { "epoch": 0.16, "learning_rate": 0.0004220060180541625, "loss": 2.9938, "theoretical_loss": 3.880643042206816, "tokens_seen": 542769152 }, { "epoch": 0.16, "learning_rate": 0.0004219959879638917, "loss": 2.7158, "theoretical_loss": 3.880593288928152, "tokens_seen": 542834688 }, { "epoch": 0.16, "learning_rate": 0.00042198595787362085, "loss": 2.9342, "theoretical_loss": 3.8805435433374353, "tokens_seen": 542900224 }, { "epoch": 0.16, "learning_rate": 0.00042197592778335004, "loss": 2.9032, "theoretical_loss": 3.880493805432552, "tokens_seen": 542965760 }, { "epoch": 0.16, "learning_rate": 0.00042196589769307927, "loss": 2.7669, "theoretical_loss": 3.8804440752113862, "tokens_seen": 543031296 }, { "epoch": 0.16, "learning_rate": 0.0004219558676028084, "loss": 2.9817, "theoretical_loss": 3.880394352671824, "tokens_seen": 543096832 }, { "epoch": 0.16, "learning_rate": 0.00042194583751253763, "loss": 2.9951, "theoretical_loss": 3.880344637811752, "tokens_seen": 543162368 }, { "epoch": 0.16, "learning_rate": 0.00042193580742226676, "loss": 3.1305, "theoretical_loss": 3.880294930629058, "tokens_seen": 543227904 }, { "epoch": 0.16, "learning_rate": 0.000421925777331996, "loss": 2.7982, "theoretical_loss": 3.8802452311216302, "tokens_seen": 543293440 }, { "epoch": 0.16, "learning_rate": 0.00042191574724172523, "loss": 2.9459, "theoretical_loss": 3.880195539287358, "tokens_seen": 543358976 }, { "epoch": 0.16, "learning_rate": 0.00042190571715145436, "loss": 2.7107, "theoretical_loss": 3.8801458551241312, "tokens_seen": 543424512 }, { "epoch": 0.16, "learning_rate": 0.0004218956870611836, "loss": 2.9262, "theoretical_loss": 3.8800961786298407, "tokens_seen": 543490048 }, { "epoch": 0.16, "learning_rate": 0.0004218856569709127, "loss": 2.9468, "theoretical_loss": 3.880046509802378, "tokens_seen": 543555584 }, { "epoch": 0.16, "learning_rate": 0.00042187562688064196, "loss": 2.8462, "theoretical_loss": 3.879996848639637, "tokens_seen": 543621120 }, { "epoch": 0.16, "learning_rate": 0.00042186559679037114, "loss": 2.8692, "theoretical_loss": 3.8799471951395095, "tokens_seen": 543686656 }, { "epoch": 0.16, "learning_rate": 0.0004218555667001003, "loss": 2.7989, "theoretical_loss": 3.8798975492998906, "tokens_seen": 543752192 }, { "epoch": 0.16, "learning_rate": 0.0004218455366098295, "loss": 2.9146, "theoretical_loss": 3.8798479111186754, "tokens_seen": 543817728 }, { "epoch": 0.16, "learning_rate": 0.00042183550651955873, "loss": 2.9114, "theoretical_loss": 3.879798280593759, "tokens_seen": 543883264 }, { "epoch": 0.16, "objective/train/docs_used": 318285, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9519379138946533, "objective/train/theoretical_loss": 3.879748657723039, "objective/train/tokens_used": 564408800, "theoretical_loss": 3.879748657723039, "tokens_seen": 543948800 }, { "epoch": 0.16, "learning_rate": 0.00042182547642928786, "loss": 2.8573, "theoretical_loss": 3.879748657723039, "tokens_seen": 543948800 }, { "epoch": 0.16, "learning_rate": 0.0004218154463390171, "loss": 2.7129, "theoretical_loss": 3.8796990425044124, "tokens_seen": 544014336 }, { "epoch": 0.16, "learning_rate": 0.0004218054162487462, "loss": 2.9859, "theoretical_loss": 3.879649434935778, "tokens_seen": 544079872 }, { "epoch": 0.16, "learning_rate": 0.00042179538615847546, "loss": 2.7631, "theoretical_loss": 3.8795998350150347, "tokens_seen": 544145408 }, { "epoch": 0.16, "learning_rate": 0.00042178535606820464, "loss": 2.7474, "theoretical_loss": 3.879550242740083, "tokens_seen": 544210944 }, { "epoch": 0.16, "learning_rate": 0.0004217753259779338, "loss": 2.88, "theoretical_loss": 3.8795006581088236, "tokens_seen": 544276480 }, { "epoch": 0.16, "learning_rate": 0.000421765295887663, "loss": 2.8279, "theoretical_loss": 3.879451081119158, "tokens_seen": 544342016 }, { "epoch": 0.16, "learning_rate": 0.0004217552657973922, "loss": 2.8613, "theoretical_loss": 3.8794015117689886, "tokens_seen": 544407552 }, { "epoch": 0.16, "learning_rate": 0.00042174523570712136, "loss": 2.9193, "theoretical_loss": 3.8793519500562192, "tokens_seen": 544473088 }, { "epoch": 0.17, "learning_rate": 0.0004217352056168506, "loss": 2.8995, "theoretical_loss": 3.879302395978754, "tokens_seen": 544538624 }, { "epoch": 0.17, "learning_rate": 0.0004217251755265797, "loss": 2.848, "theoretical_loss": 3.879252849534498, "tokens_seen": 544604160 }, { "epoch": 0.17, "learning_rate": 0.00042171514543630896, "loss": 3.0245, "theoretical_loss": 3.8792033107213566, "tokens_seen": 544669696 }, { "epoch": 0.17, "learning_rate": 0.0004217051153460381, "loss": 2.7713, "theoretical_loss": 3.8791537795372366, "tokens_seen": 544735232 }, { "epoch": 0.17, "learning_rate": 0.0004216950852557673, "loss": 2.86, "theoretical_loss": 3.879104255980046, "tokens_seen": 544800768 }, { "epoch": 0.17, "learning_rate": 0.0004216850551654965, "loss": 2.9259, "theoretical_loss": 3.8790547400476925, "tokens_seen": 544866304 }, { "epoch": 0.17, "learning_rate": 0.0004216750250752257, "loss": 2.6663, "theoretical_loss": 3.879005231738086, "tokens_seen": 544931840 }, { "epoch": 0.17, "learning_rate": 0.00042166499498495487, "loss": 2.875, "theoretical_loss": 3.878955731049136, "tokens_seen": 544997376 }, { "epoch": 0.17, "learning_rate": 0.0004216549648946841, "loss": 3.1208, "theoretical_loss": 3.878906237978753, "tokens_seen": 545062912 }, { "epoch": 0.17, "learning_rate": 0.00042164493480441323, "loss": 2.6814, "theoretical_loss": 3.8788567525248494, "tokens_seen": 545128448 }, { "epoch": 0.17, "learning_rate": 0.00042163490471414246, "loss": 2.947, "theoretical_loss": 3.878807274685337, "tokens_seen": 545193984 }, { "epoch": 0.17, "learning_rate": 0.0004216248746238716, "loss": 2.8911, "theoretical_loss": 3.8787578044581292, "tokens_seen": 545259520 }, { "epoch": 0.17, "learning_rate": 0.0004216148445336008, "loss": 2.9855, "theoretical_loss": 3.87870834184114, "tokens_seen": 545325056 }, { "epoch": 0.17, "learning_rate": 0.00042160481444333, "loss": 3.1216, "theoretical_loss": 3.8786588868322847, "tokens_seen": 545390592 }, { "epoch": 0.17, "learning_rate": 0.0004215947843530592, "loss": 2.7415, "theoretical_loss": 3.8786094394294794, "tokens_seen": 545456128 }, { "epoch": 0.17, "learning_rate": 0.00042158475426278837, "loss": 2.9065, "theoretical_loss": 3.8785599996306397, "tokens_seen": 545521664 }, { "epoch": 0.17, "objective/train/docs_used": 318961, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9253644943237305, "objective/train/theoretical_loss": 3.8785105674336835, "objective/train/tokens_used": 566047200, "theoretical_loss": 3.8785105674336835, "tokens_seen": 545587200 }, { "epoch": 0.17, "learning_rate": 0.00042157472417251755, "loss": 2.9532, "theoretical_loss": 3.8785105674336835, "tokens_seen": 545587200 }, { "epoch": 0.17, "learning_rate": 0.00042156469408224673, "loss": 3.0437, "theoretical_loss": 3.8784611428365285, "tokens_seen": 545652736 }, { "epoch": 0.17, "learning_rate": 0.00042155466399197597, "loss": 2.7187, "theoretical_loss": 3.8784117258370943, "tokens_seen": 545718272 }, { "epoch": 0.17, "learning_rate": 0.0004215446339017051, "loss": 2.7803, "theoretical_loss": 3.878362316433301, "tokens_seen": 545783808 }, { "epoch": 0.17, "learning_rate": 0.00042153460381143433, "loss": 2.9175, "theoretical_loss": 3.878312914623068, "tokens_seen": 545849344 }, { "epoch": 0.17, "learning_rate": 0.0004215245737211635, "loss": 3.0657, "theoretical_loss": 3.8782635204043188, "tokens_seen": 545914880 }, { "epoch": 0.17, "learning_rate": 0.0004215145436308927, "loss": 3.1568, "theoretical_loss": 3.8782141337749736, "tokens_seen": 545980416 }, { "epoch": 0.17, "learning_rate": 0.0004215045135406219, "loss": 3.1295, "theoretical_loss": 3.8781647547329565, "tokens_seen": 546045952 }, { "epoch": 0.17, "learning_rate": 0.00042149448345035105, "loss": 2.8782, "theoretical_loss": 3.8781153832761914, "tokens_seen": 546111488 }, { "epoch": 0.17, "learning_rate": 0.00042148445336008024, "loss": 2.8349, "theoretical_loss": 3.8780660194026035, "tokens_seen": 546177024 }, { "epoch": 0.17, "learning_rate": 0.00042147442326980947, "loss": 2.8772, "theoretical_loss": 3.8780166631101176, "tokens_seen": 546242560 }, { "epoch": 0.17, "learning_rate": 0.0004214643931795386, "loss": 2.7605, "theoretical_loss": 3.8779673143966606, "tokens_seen": 546308096 }, { "epoch": 0.17, "learning_rate": 0.00042145436308926783, "loss": 2.7632, "theoretical_loss": 3.8779179732601587, "tokens_seen": 546373632 }, { "epoch": 0.17, "learning_rate": 0.00042144433299899696, "loss": 2.7438, "theoretical_loss": 3.8778686396985416, "tokens_seen": 546439168 }, { "epoch": 0.17, "learning_rate": 0.0004214343029087262, "loss": 3.0832, "theoretical_loss": 3.877819313709736, "tokens_seen": 546504704 }, { "epoch": 0.17, "learning_rate": 0.0004214242728184554, "loss": 2.8661, "theoretical_loss": 3.8777699952916738, "tokens_seen": 546570240 }, { "epoch": 0.17, "learning_rate": 0.00042141424272818456, "loss": 2.748, "theoretical_loss": 3.8777206844422842, "tokens_seen": 546635776 }, { "epoch": 0.17, "learning_rate": 0.00042140421263791374, "loss": 2.7212, "theoretical_loss": 3.877671381159499, "tokens_seen": 546701312 }, { "epoch": 0.17, "learning_rate": 0.0004213941825476429, "loss": 2.6799, "theoretical_loss": 3.8776220854412493, "tokens_seen": 546766848 }, { "epoch": 0.17, "learning_rate": 0.0004213841524573721, "loss": 2.8845, "theoretical_loss": 3.877572797285469, "tokens_seen": 546832384 }, { "epoch": 0.17, "learning_rate": 0.00042137412236710134, "loss": 3.0905, "theoretical_loss": 3.8775235166900917, "tokens_seen": 546897920 }, { "epoch": 0.17, "learning_rate": 0.00042136409227683046, "loss": 2.9857, "theoretical_loss": 3.877474243653051, "tokens_seen": 546963456 }, { "epoch": 0.17, "learning_rate": 0.0004213540621865597, "loss": 2.8874, "theoretical_loss": 3.8774249781722836, "tokens_seen": 547028992 }, { "epoch": 0.17, "learning_rate": 0.0004213440320962889, "loss": 2.9957, "theoretical_loss": 3.877375720245725, "tokens_seen": 547094528 }, { "epoch": 0.17, "learning_rate": 0.00042133400200601806, "loss": 2.8372, "theoretical_loss": 3.8773264698713117, "tokens_seen": 547160064 }, { "epoch": 0.17, "objective/train/docs_used": 320353, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.186674118041992, "objective/train/theoretical_loss": 3.8772772270469824, "objective/train/tokens_used": 567685600, "theoretical_loss": 3.8772772270469824, "tokens_seen": 547225600 }, { "epoch": 0.17, "learning_rate": 0.00042132397191574724, "loss": 3.0207, "theoretical_loss": 3.8772772270469824, "tokens_seen": 547225600 }, { "epoch": 0.17, "learning_rate": 0.0004213139418254764, "loss": 2.9139, "theoretical_loss": 3.877227991770675, "tokens_seen": 547291136 }, { "epoch": 0.17, "learning_rate": 0.0004213039117352056, "loss": 2.9367, "theoretical_loss": 3.877178764040329, "tokens_seen": 547356672 }, { "epoch": 0.17, "learning_rate": 0.00042129388164493484, "loss": 2.9502, "theoretical_loss": 3.8771295438538846, "tokens_seen": 547422208 }, { "epoch": 0.17, "learning_rate": 0.00042128385155466397, "loss": 2.9549, "theoretical_loss": 3.8770803312092834, "tokens_seen": 547487744 }, { "epoch": 0.17, "learning_rate": 0.0004212738214643932, "loss": 2.8362, "theoretical_loss": 3.877031126104466, "tokens_seen": 547553280 }, { "epoch": 0.17, "learning_rate": 0.00042126379137412233, "loss": 2.7986, "theoretical_loss": 3.8769819285373766, "tokens_seen": 547618816 }, { "epoch": 0.17, "learning_rate": 0.00042125376128385156, "loss": 2.8222, "theoretical_loss": 3.876932738505957, "tokens_seen": 547684352 }, { "epoch": 0.17, "learning_rate": 0.00042124373119358075, "loss": 2.9447, "theoretical_loss": 3.8768835560081527, "tokens_seen": 547749888 }, { "epoch": 0.17, "learning_rate": 0.0004212337011033099, "loss": 2.7848, "theoretical_loss": 3.876834381041908, "tokens_seen": 547815424 }, { "epoch": 0.17, "learning_rate": 0.0004212236710130391, "loss": 2.8295, "theoretical_loss": 3.8767852136051686, "tokens_seen": 547880960 }, { "epoch": 0.17, "learning_rate": 0.0004212136409227683, "loss": 2.7802, "theoretical_loss": 3.8767360536958817, "tokens_seen": 547946496 }, { "epoch": 0.17, "learning_rate": 0.00042120361083249747, "loss": 2.8975, "theoretical_loss": 3.876686901311995, "tokens_seen": 548012032 }, { "epoch": 0.17, "learning_rate": 0.0004211935807422267, "loss": 2.9494, "theoretical_loss": 3.8766377564514563, "tokens_seen": 548077568 }, { "epoch": 0.17, "learning_rate": 0.00042118355065195583, "loss": 2.7224, "theoretical_loss": 3.8765886191122148, "tokens_seen": 548143104 }, { "epoch": 0.17, "learning_rate": 0.00042117352056168507, "loss": 2.9374, "theoretical_loss": 3.87653948929222, "tokens_seen": 548208640 }, { "epoch": 0.17, "learning_rate": 0.0004211634904714143, "loss": 3.0219, "theoretical_loss": 3.876490366989423, "tokens_seen": 548274176 }, { "epoch": 0.17, "learning_rate": 0.00042115346038114343, "loss": 2.974, "theoretical_loss": 3.876441252201775, "tokens_seen": 548339712 }, { "epoch": 0.17, "learning_rate": 0.00042114343029087266, "loss": 2.9682, "theoretical_loss": 3.876392144927229, "tokens_seen": 548405248 }, { "epoch": 0.17, "learning_rate": 0.0004211334002006018, "loss": 2.8559, "theoretical_loss": 3.876343045163737, "tokens_seen": 548470784 }, { "epoch": 0.17, "learning_rate": 0.00042112337011033103, "loss": 3.0035, "theoretical_loss": 3.8762939529092537, "tokens_seen": 548536320 }, { "epoch": 0.17, "learning_rate": 0.0004211133400200602, "loss": 2.7531, "theoretical_loss": 3.8762448681617334, "tokens_seen": 548601856 }, { "epoch": 0.17, "learning_rate": 0.0004211033099297894, "loss": 2.7791, "theoretical_loss": 3.876195790919132, "tokens_seen": 548667392 }, { "epoch": 0.17, "learning_rate": 0.00042109327983951857, "loss": 2.74, "theoretical_loss": 3.876146721179405, "tokens_seen": 548732928 }, { "epoch": 0.17, "learning_rate": 0.00042108324974924775, "loss": 2.8013, "theoretical_loss": 3.8760976589405107, "tokens_seen": 548798464 }, { "epoch": 0.17, "objective/train/docs_used": 320989, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.815619945526123, "objective/train/theoretical_loss": 3.8760486042004056, "objective/train/tokens_used": 569324000, "theoretical_loss": 3.8760486042004056, "tokens_seen": 548864000 }, { "epoch": 0.17, "learning_rate": 0.00042107321965897693, "loss": 2.6059, "theoretical_loss": 3.8760486042004056, "tokens_seen": 548864000 }, { "epoch": 0.17, "learning_rate": 0.00042106318956870617, "loss": 2.8879, "theoretical_loss": 3.8759995569570496, "tokens_seen": 548929536 }, { "epoch": 0.17, "learning_rate": 0.0004210531594784353, "loss": 2.746, "theoretical_loss": 3.8759505172084014, "tokens_seen": 548995072 }, { "epoch": 0.17, "learning_rate": 0.00042104312938816453, "loss": 2.9174, "theoretical_loss": 3.875901484952422, "tokens_seen": 549060608 }, { "epoch": 0.17, "learning_rate": 0.0004210330992978937, "loss": 2.8429, "theoretical_loss": 3.8758524601870716, "tokens_seen": 549126144 }, { "epoch": 0.17, "learning_rate": 0.0004210230692076229, "loss": 2.7353, "theoretical_loss": 3.8758034429103128, "tokens_seen": 549191680 }, { "epoch": 0.17, "learning_rate": 0.0004210130391173521, "loss": 2.6284, "theoretical_loss": 3.8757544331201084, "tokens_seen": 549257216 }, { "epoch": 0.17, "learning_rate": 0.00042100300902708125, "loss": 2.7378, "theoretical_loss": 3.875705430814421, "tokens_seen": 549322752 }, { "epoch": 0.17, "learning_rate": 0.00042099297893681044, "loss": 2.729, "theoretical_loss": 3.8756564359912153, "tokens_seen": 549388288 }, { "epoch": 0.17, "learning_rate": 0.00042098294884653967, "loss": 2.7258, "theoretical_loss": 3.8756074486484566, "tokens_seen": 549453824 }, { "epoch": 0.17, "learning_rate": 0.0004209729187562688, "loss": 2.7642, "theoretical_loss": 3.8755584687841114, "tokens_seen": 549519360 }, { "epoch": 0.17, "learning_rate": 0.00042096288866599803, "loss": 2.6261, "theoretical_loss": 3.8755094963961447, "tokens_seen": 549584896 }, { "epoch": 0.17, "learning_rate": 0.00042095285857572716, "loss": 2.6199, "theoretical_loss": 3.875460531482526, "tokens_seen": 549650432 }, { "epoch": 0.17, "learning_rate": 0.0004209428284854564, "loss": 2.8711, "theoretical_loss": 3.8754115740412214, "tokens_seen": 549715968 }, { "epoch": 0.17, "learning_rate": 0.0004209327983951856, "loss": 2.8933, "theoretical_loss": 3.8753626240702013, "tokens_seen": 549781504 }, { "epoch": 0.17, "learning_rate": 0.00042092276830491476, "loss": 2.8939, "theoretical_loss": 3.8753136815674356, "tokens_seen": 549847040 }, { "epoch": 0.17, "learning_rate": 0.00042091273821464394, "loss": 2.6116, "theoretical_loss": 3.875264746530894, "tokens_seen": 549912576 }, { "epoch": 0.17, "learning_rate": 0.0004209027081243731, "loss": 3.0494, "theoretical_loss": 3.875215818958549, "tokens_seen": 549978112 }, { "epoch": 0.17, "learning_rate": 0.0004208926780341023, "loss": 2.8988, "theoretical_loss": 3.875166898848372, "tokens_seen": 550043648 }, { "epoch": 0.17, "learning_rate": 0.00042088264794383154, "loss": 2.7881, "theoretical_loss": 3.8751179861983367, "tokens_seen": 550109184 }, { "epoch": 0.17, "learning_rate": 0.00042087261785356066, "loss": 2.8783, "theoretical_loss": 3.8750690810064166, "tokens_seen": 550174720 }, { "epoch": 0.17, "learning_rate": 0.0004208625877632899, "loss": 2.7529, "theoretical_loss": 3.8750201832705864, "tokens_seen": 550240256 }, { "epoch": 0.17, "learning_rate": 0.0004208525576730191, "loss": 2.9125, "theoretical_loss": 3.874971292988821, "tokens_seen": 550305792 }, { "epoch": 0.17, "learning_rate": 0.00042084252758274826, "loss": 2.791, "theoretical_loss": 3.8749224101590976, "tokens_seen": 550371328 }, { "epoch": 0.17, "learning_rate": 0.00042083249749247744, "loss": 3.0134, "theoretical_loss": 3.8748735347793923, "tokens_seen": 550436864 }, { "epoch": 0.17, "objective/train/docs_used": 322374, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5075249671936035, "objective/train/theoretical_loss": 3.8748246668476827, "objective/train/tokens_used": 570962400, "theoretical_loss": 3.8748246668476827, "tokens_seen": 550502400 }, { "epoch": 0.17, "learning_rate": 0.0004208224674022066, "loss": 2.8429, "theoretical_loss": 3.8748246668476827, "tokens_seen": 550502400 }, { "epoch": 0.17, "learning_rate": 0.0004208124373119358, "loss": 2.8469, "theoretical_loss": 3.874775806361948, "tokens_seen": 550567936 }, { "epoch": 0.17, "learning_rate": 0.00042080240722166504, "loss": 2.9299, "theoretical_loss": 3.8747269533201676, "tokens_seen": 550633472 }, { "epoch": 0.17, "learning_rate": 0.00042079237713139417, "loss": 2.8438, "theoretical_loss": 3.874678107720321, "tokens_seen": 550699008 }, { "epoch": 0.17, "learning_rate": 0.0004207823470411234, "loss": 2.8016, "theoretical_loss": 3.8746292695603897, "tokens_seen": 550764544 }, { "epoch": 0.17, "learning_rate": 0.00042077231695085253, "loss": 2.9375, "theoretical_loss": 3.874580438838355, "tokens_seen": 550830080 }, { "epoch": 0.17, "learning_rate": 0.00042076228686058176, "loss": 2.9417, "theoretical_loss": 3.8745316155521996, "tokens_seen": 550895616 }, { "epoch": 0.17, "learning_rate": 0.00042075225677031095, "loss": 3.0751, "theoretical_loss": 3.8744827996999067, "tokens_seen": 550961152 }, { "epoch": 0.17, "learning_rate": 0.0004207422266800401, "loss": 2.6153, "theoretical_loss": 3.874433991279461, "tokens_seen": 551026688 }, { "epoch": 0.17, "learning_rate": 0.0004207321965897693, "loss": 2.8295, "theoretical_loss": 3.874385190288846, "tokens_seen": 551092224 }, { "epoch": 0.17, "learning_rate": 0.0004207221664994985, "loss": 2.5518, "theoretical_loss": 3.874336396726049, "tokens_seen": 551157760 }, { "epoch": 0.17, "learning_rate": 0.00042071213640922767, "loss": 2.8555, "theoretical_loss": 3.8742876105890547, "tokens_seen": 551223296 }, { "epoch": 0.17, "learning_rate": 0.0004207021063189569, "loss": 2.9313, "theoretical_loss": 3.874238831875852, "tokens_seen": 551288832 }, { "epoch": 0.17, "learning_rate": 0.00042069207622868603, "loss": 2.8664, "theoretical_loss": 3.8741900605844273, "tokens_seen": 551354368 }, { "epoch": 0.17, "learning_rate": 0.00042068204613841527, "loss": 2.6718, "theoretical_loss": 3.8741412967127706, "tokens_seen": 551419904 }, { "epoch": 0.17, "learning_rate": 0.00042067201604814445, "loss": 2.9157, "theoretical_loss": 3.8740925402588706, "tokens_seen": 551485440 }, { "epoch": 0.17, "learning_rate": 0.00042066198595787363, "loss": 2.8645, "theoretical_loss": 3.8740437912207186, "tokens_seen": 551550976 }, { "epoch": 0.17, "learning_rate": 0.0004206519558676028, "loss": 2.8228, "theoretical_loss": 3.873995049596305, "tokens_seen": 551616512 }, { "epoch": 0.17, "learning_rate": 0.000420641925777332, "loss": 3.0906, "theoretical_loss": 3.873946315383622, "tokens_seen": 551682048 }, { "epoch": 0.17, "learning_rate": 0.0004206318956870612, "loss": 2.8373, "theoretical_loss": 3.873897588580662, "tokens_seen": 551747584 }, { "epoch": 0.17, "learning_rate": 0.0004206218655967904, "loss": 2.767, "theoretical_loss": 3.8738488691854194, "tokens_seen": 551813120 }, { "epoch": 0.17, "learning_rate": 0.00042061183550651954, "loss": 3.1444, "theoretical_loss": 3.873800157195887, "tokens_seen": 551878656 }, { "epoch": 0.17, "learning_rate": 0.00042060180541624877, "loss": 2.8976, "theoretical_loss": 3.873751452610061, "tokens_seen": 551944192 }, { "epoch": 0.17, "learning_rate": 0.0004205917753259779, "loss": 2.717, "theoretical_loss": 3.8737027554259367, "tokens_seen": 552009728 }, { "epoch": 0.17, "learning_rate": 0.00042058174523570713, "loss": 2.8454, "theoretical_loss": 3.873654065641511, "tokens_seen": 552075264 }, { "epoch": 0.17, "objective/train/docs_used": 322945, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0310089588165283, "objective/train/theoretical_loss": 3.8736053832547808, "objective/train/tokens_used": 572600800, "theoretical_loss": 3.8736053832547808, "tokens_seen": 552140800 }, { "epoch": 0.17, "learning_rate": 0.0004205717151454363, "loss": 2.8704, "theoretical_loss": 3.8736053832547808, "tokens_seen": 552140800 }, { "epoch": 0.17, "learning_rate": 0.0004205616850551655, "loss": 2.6181, "theoretical_loss": 3.873556708263745, "tokens_seen": 552206336 }, { "epoch": 0.17, "learning_rate": 0.0004205516549648947, "loss": 2.9212, "theoretical_loss": 3.873508040666402, "tokens_seen": 552271872 }, { "epoch": 0.17, "learning_rate": 0.0004205416248746239, "loss": 2.9482, "theoretical_loss": 3.8734593804607513, "tokens_seen": 552337408 }, { "epoch": 0.17, "learning_rate": 0.00042053159478435304, "loss": 2.789, "theoretical_loss": 3.873410727644794, "tokens_seen": 552402944 }, { "epoch": 0.17, "learning_rate": 0.0004205215646940823, "loss": 3.0142, "theoretical_loss": 3.8733620822165316, "tokens_seen": 552468480 }, { "epoch": 0.17, "learning_rate": 0.0004205115346038114, "loss": 2.9676, "theoretical_loss": 3.8733134441739647, "tokens_seen": 552534016 }, { "epoch": 0.17, "learning_rate": 0.00042050150451354064, "loss": 2.8231, "theoretical_loss": 3.8732648135150978, "tokens_seen": 552599552 }, { "epoch": 0.17, "learning_rate": 0.0004204914744232698, "loss": 2.862, "theoretical_loss": 3.8732161902379336, "tokens_seen": 552665088 }, { "epoch": 0.17, "learning_rate": 0.000420481444332999, "loss": 2.9855, "theoretical_loss": 3.873167574340477, "tokens_seen": 552730624 }, { "epoch": 0.17, "learning_rate": 0.0004204714142427282, "loss": 2.7015, "theoretical_loss": 3.873118965820733, "tokens_seen": 552796160 }, { "epoch": 0.17, "learning_rate": 0.00042046138415245736, "loss": 2.8132, "theoretical_loss": 3.873070364676707, "tokens_seen": 552861696 }, { "epoch": 0.17, "learning_rate": 0.00042045135406218654, "loss": 2.8356, "theoretical_loss": 3.8730217709064063, "tokens_seen": 552927232 }, { "epoch": 0.17, "learning_rate": 0.0004204413239719158, "loss": 2.9976, "theoretical_loss": 3.8729731845078383, "tokens_seen": 552992768 }, { "epoch": 0.17, "learning_rate": 0.0004204312938816449, "loss": 3.1535, "theoretical_loss": 3.872924605479011, "tokens_seen": 553058304 }, { "epoch": 0.17, "learning_rate": 0.00042042126379137414, "loss": 2.526, "theoretical_loss": 3.872876033817934, "tokens_seen": 553123840 }, { "epoch": 0.17, "learning_rate": 0.0004204112337011033, "loss": 2.8647, "theoretical_loss": 3.8728274695226164, "tokens_seen": 553189376 }, { "epoch": 0.17, "learning_rate": 0.0004204012036108325, "loss": 3.1751, "theoretical_loss": 3.872778912591069, "tokens_seen": 553254912 }, { "epoch": 0.17, "learning_rate": 0.00042039117352056174, "loss": 2.7206, "theoretical_loss": 3.872730363021304, "tokens_seen": 553320448 }, { "epoch": 0.17, "learning_rate": 0.00042038114343029086, "loss": 2.7845, "theoretical_loss": 3.8726818208113323, "tokens_seen": 553385984 }, { "epoch": 0.17, "learning_rate": 0.0004203711133400201, "loss": 3.2867, "theoretical_loss": 3.872633285959167, "tokens_seen": 553451520 }, { "epoch": 0.17, "learning_rate": 0.0004203610832497493, "loss": 2.8596, "theoretical_loss": 3.872584758462823, "tokens_seen": 553517056 }, { "epoch": 0.17, "learning_rate": 0.00042035105315947846, "loss": 2.7029, "theoretical_loss": 3.872536238320313, "tokens_seen": 553582592 }, { "epoch": 0.17, "learning_rate": 0.00042034102306920764, "loss": 2.7705, "theoretical_loss": 3.8724877255296533, "tokens_seen": 553648128 }, { "epoch": 0.17, "learning_rate": 0.0004203309929789368, "loss": 2.9573, "theoretical_loss": 3.872439220088859, "tokens_seen": 553713664 }, { "epoch": 0.17, "objective/train/docs_used": 324208, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7510035037994385, "objective/train/theoretical_loss": 3.8723907219959486, "objective/train/tokens_used": 574239200, "theoretical_loss": 3.8723907219959486, "tokens_seen": 553779200 }, { "epoch": 0.17, "learning_rate": 0.000420320962888666, "loss": 2.7948, "theoretical_loss": 3.8723907219959486, "tokens_seen": 553779200 }, { "epoch": 0.17, "learning_rate": 0.00042031093279839524, "loss": 3.0678, "theoretical_loss": 3.872342231248938, "tokens_seen": 553844736 }, { "epoch": 0.17, "learning_rate": 0.00042030090270812437, "loss": 2.913, "theoretical_loss": 3.872293747845846, "tokens_seen": 553910272 }, { "epoch": 0.17, "learning_rate": 0.0004202908726178536, "loss": 2.8919, "theoretical_loss": 3.872245271784692, "tokens_seen": 553975808 }, { "epoch": 0.17, "learning_rate": 0.00042028084252758273, "loss": 2.9291, "theoretical_loss": 3.872196803063495, "tokens_seen": 554041344 }, { "epoch": 0.17, "learning_rate": 0.00042027081243731196, "loss": 2.8092, "theoretical_loss": 3.872148341680276, "tokens_seen": 554106880 }, { "epoch": 0.17, "learning_rate": 0.00042026078234704115, "loss": 2.8927, "theoretical_loss": 3.8720998876330572, "tokens_seen": 554172416 }, { "epoch": 0.17, "learning_rate": 0.0004202507522567703, "loss": 2.9609, "theoretical_loss": 3.8720514409198596, "tokens_seen": 554237952 }, { "epoch": 0.17, "learning_rate": 0.0004202407221664995, "loss": 2.6144, "theoretical_loss": 3.8720030015387064, "tokens_seen": 554303488 }, { "epoch": 0.17, "learning_rate": 0.0004202306920762287, "loss": 3.0194, "theoretical_loss": 3.871954569487622, "tokens_seen": 554369024 }, { "epoch": 0.17, "learning_rate": 0.00042022066198595787, "loss": 2.7512, "theoretical_loss": 3.8719061447646297, "tokens_seen": 554434560 }, { "epoch": 0.17, "learning_rate": 0.0004202106318956871, "loss": 2.496, "theoretical_loss": 3.8718577273677557, "tokens_seen": 554500096 }, { "epoch": 0.17, "learning_rate": 0.00042020060180541623, "loss": 2.7344, "theoretical_loss": 3.8718093172950248, "tokens_seen": 554565632 }, { "epoch": 0.17, "learning_rate": 0.00042019057171514547, "loss": 2.8702, "theoretical_loss": 3.8717609145444647, "tokens_seen": 554631168 }, { "epoch": 0.17, "learning_rate": 0.00042018054162487465, "loss": 2.7379, "theoretical_loss": 3.8717125191141024, "tokens_seen": 554696704 }, { "epoch": 0.17, "learning_rate": 0.00042017051153460383, "loss": 2.8682, "theoretical_loss": 3.871664131001967, "tokens_seen": 554762240 }, { "epoch": 0.17, "learning_rate": 0.000420160481444333, "loss": 2.6698, "theoretical_loss": 3.8716157502060864, "tokens_seen": 554827776 }, { "epoch": 0.17, "learning_rate": 0.0004201504513540622, "loss": 3.04, "theoretical_loss": 3.8715673767244914, "tokens_seen": 554893312 }, { "epoch": 0.17, "learning_rate": 0.0004201404212637914, "loss": 2.9472, "theoretical_loss": 3.871519010555211, "tokens_seen": 554958848 }, { "epoch": 0.17, "learning_rate": 0.0004201303911735206, "loss": 2.8381, "theoretical_loss": 3.8714706516962787, "tokens_seen": 555024384 }, { "epoch": 0.17, "learning_rate": 0.00042012036108324974, "loss": 2.975, "theoretical_loss": 3.871422300145725, "tokens_seen": 555089920 }, { "epoch": 0.17, "learning_rate": 0.00042011033099297897, "loss": 2.7581, "theoretical_loss": 3.8713739559015834, "tokens_seen": 555155456 }, { "epoch": 0.17, "learning_rate": 0.0004201003009027081, "loss": 2.7519, "theoretical_loss": 3.871325618961887, "tokens_seen": 555220992 }, { "epoch": 0.17, "learning_rate": 0.00042009027081243733, "loss": 2.8407, "theoretical_loss": 3.871277289324671, "tokens_seen": 555286528 }, { "epoch": 0.17, "learning_rate": 0.0004200802407221665, "loss": 3.044, "theoretical_loss": 3.871228966987969, "tokens_seen": 555352064 }, { "epoch": 0.17, "objective/train/docs_used": 324930, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.34468674659729, "objective/train/theoretical_loss": 3.871180651949819, "objective/train/tokens_used": 575877600, "theoretical_loss": 3.871180651949819, "tokens_seen": 555417600 }, { "epoch": 0.17, "learning_rate": 0.0004200702106318957, "loss": 2.9225, "theoretical_loss": 3.871180651949819, "tokens_seen": 555417600 }, { "epoch": 0.17, "learning_rate": 0.0004200601805416249, "loss": 2.9334, "theoretical_loss": 3.871132344208256, "tokens_seen": 555483136 }, { "epoch": 0.17, "learning_rate": 0.0004200501504513541, "loss": 2.6462, "theoretical_loss": 3.871084043761318, "tokens_seen": 555548672 }, { "epoch": 0.17, "learning_rate": 0.00042004012036108324, "loss": 3.0531, "theoretical_loss": 3.8710357506070427, "tokens_seen": 555614208 }, { "epoch": 0.17, "learning_rate": 0.0004200300902708125, "loss": 3.0704, "theoretical_loss": 3.87098746474347, "tokens_seen": 555679744 }, { "epoch": 0.17, "learning_rate": 0.0004200200601805416, "loss": 2.768, "theoretical_loss": 3.870939186168638, "tokens_seen": 555745280 }, { "epoch": 0.17, "learning_rate": 0.00042001003009027084, "loss": 2.888, "theoretical_loss": 3.8708909148805892, "tokens_seen": 555810816 }, { "epoch": 0.17, "learning_rate": 0.00042, "loss": 2.9139, "theoretical_loss": 3.8708426508773632, "tokens_seen": 555876352 }, { "epoch": 0.17, "learning_rate": 0.0004199899699097292, "loss": 2.7703, "theoretical_loss": 3.870794394157002, "tokens_seen": 555941888 }, { "epoch": 0.17, "learning_rate": 0.0004199799398194584, "loss": 2.8197, "theoretical_loss": 3.8707461447175495, "tokens_seen": 556007424 }, { "epoch": 0.17, "learning_rate": 0.00041996990972918756, "loss": 2.6741, "theoretical_loss": 3.8706979025570485, "tokens_seen": 556072960 }, { "epoch": 0.17, "learning_rate": 0.00041995987963891674, "loss": 2.7795, "theoretical_loss": 3.870649667673543, "tokens_seen": 556138496 }, { "epoch": 0.17, "learning_rate": 0.000419949849548646, "loss": 2.8733, "theoretical_loss": 3.870601440065078, "tokens_seen": 556204032 }, { "epoch": 0.17, "learning_rate": 0.0004199398194583751, "loss": 2.619, "theoretical_loss": 3.8705532197296995, "tokens_seen": 556269568 }, { "epoch": 0.17, "learning_rate": 0.00041992978936810434, "loss": 2.6645, "theoretical_loss": 3.8705050066654536, "tokens_seen": 556335104 }, { "epoch": 0.17, "learning_rate": 0.00041991975927783347, "loss": 2.9262, "theoretical_loss": 3.870456800870388, "tokens_seen": 556400640 }, { "epoch": 0.17, "learning_rate": 0.0004199097291875627, "loss": 2.6335, "theoretical_loss": 3.8704086023425512, "tokens_seen": 556466176 }, { "epoch": 0.17, "learning_rate": 0.0004198996990972919, "loss": 2.961, "theoretical_loss": 3.8703604110799903, "tokens_seen": 556531712 }, { "epoch": 0.17, "learning_rate": 0.00041988966900702106, "loss": 2.8548, "theoretical_loss": 3.870312227080756, "tokens_seen": 556597248 }, { "epoch": 0.17, "learning_rate": 0.00041987963891675025, "loss": 2.7521, "theoretical_loss": 3.8702640503428984, "tokens_seen": 556662784 }, { "epoch": 0.17, "learning_rate": 0.0004198696088264795, "loss": 2.8326, "theoretical_loss": 3.870215880864469, "tokens_seen": 556728320 }, { "epoch": 0.17, "learning_rate": 0.0004198595787362086, "loss": 2.9862, "theoretical_loss": 3.8701677186435184, "tokens_seen": 556793856 }, { "epoch": 0.17, "learning_rate": 0.00041984954864593784, "loss": 2.7987, "theoretical_loss": 3.8701195636781, "tokens_seen": 556859392 }, { "epoch": 0.17, "learning_rate": 0.00041983951855566697, "loss": 2.9304, "theoretical_loss": 3.8700714159662666, "tokens_seen": 556924928 }, { "epoch": 0.17, "learning_rate": 0.0004198294884653962, "loss": 2.8822, "theoretical_loss": 3.870023275506073, "tokens_seen": 556990464 }, { "epoch": 0.17, "objective/train/docs_used": 326252, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7411465644836426, "objective/train/theoretical_loss": 3.869975142295573, "objective/train/tokens_used": 577516000, "theoretical_loss": 3.869975142295573, "tokens_seen": 557056000 }, { "epoch": 0.17, "learning_rate": 0.0004198194583751254, "loss": 2.9452, "theoretical_loss": 3.869975142295573, "tokens_seen": 557056000 }, { "epoch": 0.17, "learning_rate": 0.00041980942828485457, "loss": 2.7595, "theoretical_loss": 3.8699270163328223, "tokens_seen": 557121536 }, { "epoch": 0.17, "learning_rate": 0.00041979939819458375, "loss": 2.8155, "theoretical_loss": 3.8698788976158776, "tokens_seen": 557187072 }, { "epoch": 0.17, "learning_rate": 0.00041978936810431293, "loss": 2.7187, "theoretical_loss": 3.8698307861427956, "tokens_seen": 557252608 }, { "epoch": 0.17, "learning_rate": 0.0004197793380140421, "loss": 2.9983, "theoretical_loss": 3.8697826819116345, "tokens_seen": 557318144 }, { "epoch": 0.17, "learning_rate": 0.00041976930792377135, "loss": 2.743, "theoretical_loss": 3.8697345849204523, "tokens_seen": 557383680 }, { "epoch": 0.17, "learning_rate": 0.00041975927783350047, "loss": 2.8873, "theoretical_loss": 3.8696864951673087, "tokens_seen": 557449216 }, { "epoch": 0.17, "learning_rate": 0.0004197492477432297, "loss": 2.7194, "theoretical_loss": 3.8696384126502634, "tokens_seen": 557514752 }, { "epoch": 0.17, "learning_rate": 0.00041973921765295883, "loss": 2.9329, "theoretical_loss": 3.8695903373673772, "tokens_seen": 557580288 }, { "epoch": 0.17, "learning_rate": 0.00041972918756268807, "loss": 2.825, "theoretical_loss": 3.869542269316711, "tokens_seen": 557645824 }, { "epoch": 0.17, "learning_rate": 0.00041971915747241725, "loss": 2.9068, "theoretical_loss": 3.869494208496329, "tokens_seen": 557711360 }, { "epoch": 0.17, "learning_rate": 0.00041970912738214643, "loss": 2.8372, "theoretical_loss": 3.8694461549042924, "tokens_seen": 557776896 }, { "epoch": 0.17, "learning_rate": 0.0004196990972918756, "loss": 3.0069, "theoretical_loss": 3.8693981085386655, "tokens_seen": 557842432 }, { "epoch": 0.17, "learning_rate": 0.00041968906720160485, "loss": 2.7972, "theoretical_loss": 3.8693500693975125, "tokens_seen": 557907968 }, { "epoch": 0.17, "learning_rate": 0.000419679037111334, "loss": 2.6788, "theoretical_loss": 3.8693020374788993, "tokens_seen": 557973504 }, { "epoch": 0.17, "learning_rate": 0.0004196690070210632, "loss": 2.8561, "theoretical_loss": 3.8692540127808916, "tokens_seen": 558039040 }, { "epoch": 0.17, "learning_rate": 0.0004196589769307924, "loss": 2.5665, "theoretical_loss": 3.8692059953015563, "tokens_seen": 558104576 }, { "epoch": 0.17, "learning_rate": 0.0004196489468405216, "loss": 2.8662, "theoretical_loss": 3.86915798503896, "tokens_seen": 558170112 }, { "epoch": 0.17, "learning_rate": 0.0004196389167502508, "loss": 3.04, "theoretical_loss": 3.8691099819911723, "tokens_seen": 558235648 }, { "epoch": 0.17, "learning_rate": 0.00041962888665997994, "loss": 2.6662, "theoretical_loss": 3.869061986156261, "tokens_seen": 558301184 }, { "epoch": 0.17, "learning_rate": 0.00041961885656970917, "loss": 2.9142, "theoretical_loss": 3.8690139975322966, "tokens_seen": 558366720 }, { "epoch": 0.17, "learning_rate": 0.0004196088264794383, "loss": 2.968, "theoretical_loss": 3.8689660161173487, "tokens_seen": 558432256 }, { "epoch": 0.17, "learning_rate": 0.00041959879638916753, "loss": 2.9044, "theoretical_loss": 3.8689180419094895, "tokens_seen": 558497792 }, { "epoch": 0.17, "learning_rate": 0.0004195887662988967, "loss": 2.7042, "theoretical_loss": 3.8688700749067904, "tokens_seen": 558563328 }, { "epoch": 0.17, "learning_rate": 0.0004195787362086259, "loss": 2.8081, "theoretical_loss": 3.868822115107324, "tokens_seen": 558628864 }, { "epoch": 0.17, "objective/train/docs_used": 326886, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5210978984832764, "objective/train/theoretical_loss": 3.8687741625091636, "objective/train/tokens_used": 579154400, "theoretical_loss": 3.8687741625091636, "tokens_seen": 558694400 }, { "epoch": 0.17, "learning_rate": 0.0004195687061183551, "loss": 2.6311, "theoretical_loss": 3.8687741625091636, "tokens_seen": 558694400 }, { "epoch": 0.17, "learning_rate": 0.0004195586760280843, "loss": 2.9603, "theoretical_loss": 3.8687262171103836, "tokens_seen": 558759936 }, { "epoch": 0.17, "learning_rate": 0.00041954864593781344, "loss": 2.8054, "theoretical_loss": 3.8686782789090595, "tokens_seen": 558825472 }, { "epoch": 0.17, "learning_rate": 0.0004195386158475427, "loss": 2.7241, "theoretical_loss": 3.868630347903266, "tokens_seen": 558891008 }, { "epoch": 0.17, "learning_rate": 0.0004195285857572718, "loss": 2.7745, "theoretical_loss": 3.8685824240910804, "tokens_seen": 558956544 }, { "epoch": 0.17, "learning_rate": 0.00041951855566700104, "loss": 2.7074, "theoretical_loss": 3.868534507470579, "tokens_seen": 559022080 }, { "epoch": 0.17, "learning_rate": 0.0004195085255767302, "loss": 2.8136, "theoretical_loss": 3.8684865980398393, "tokens_seen": 559087616 }, { "epoch": 0.17, "learning_rate": 0.0004194984954864594, "loss": 2.8338, "theoretical_loss": 3.868438695796941, "tokens_seen": 559153152 }, { "epoch": 0.17, "learning_rate": 0.0004194884653961886, "loss": 2.8892, "theoretical_loss": 3.8683908007399634, "tokens_seen": 559218688 }, { "epoch": 0.17, "learning_rate": 0.00041947843530591776, "loss": 3.0437, "theoretical_loss": 3.8683429128669857, "tokens_seen": 559284224 }, { "epoch": 0.17, "learning_rate": 0.00041946840521564694, "loss": 2.8545, "theoretical_loss": 3.8682950321760896, "tokens_seen": 559349760 }, { "epoch": 0.17, "learning_rate": 0.0004194583751253762, "loss": 2.7766, "theoretical_loss": 3.868247158665356, "tokens_seen": 559415296 }, { "epoch": 0.17, "learning_rate": 0.0004194483450351053, "loss": 2.9612, "theoretical_loss": 3.8681992923328674, "tokens_seen": 559480832 }, { "epoch": 0.17, "learning_rate": 0.00041943831494483454, "loss": 2.658, "theoretical_loss": 3.868151433176707, "tokens_seen": 559546368 }, { "epoch": 0.17, "learning_rate": 0.00041942828485456367, "loss": 2.6934, "theoretical_loss": 3.8681035811949585, "tokens_seen": 559611904 }, { "epoch": 0.17, "learning_rate": 0.0004194182547642929, "loss": 2.6309, "theoretical_loss": 3.868055736385706, "tokens_seen": 559677440 }, { "epoch": 0.17, "learning_rate": 0.0004194082246740221, "loss": 2.9048, "theoretical_loss": 3.868007898747036, "tokens_seen": 559742976 }, { "epoch": 0.17, "learning_rate": 0.00041939819458375126, "loss": 2.7095, "theoretical_loss": 3.867960068277033, "tokens_seen": 559808512 }, { "epoch": 0.17, "learning_rate": 0.00041938816449348045, "loss": 2.8633, "theoretical_loss": 3.867912244973784, "tokens_seen": 559874048 }, { "epoch": 0.17, "learning_rate": 0.0004193781344032097, "loss": 3.0507, "theoretical_loss": 3.867864428835377, "tokens_seen": 559939584 }, { "epoch": 0.17, "learning_rate": 0.0004193681043129388, "loss": 2.9638, "theoretical_loss": 3.8678166198598998, "tokens_seen": 560005120 }, { "epoch": 0.17, "learning_rate": 0.00041935807422266804, "loss": 2.8753, "theoretical_loss": 3.8677688180454415, "tokens_seen": 560070656 }, { "epoch": 0.17, "learning_rate": 0.00041934804413239717, "loss": 2.7619, "theoretical_loss": 3.867721023390091, "tokens_seen": 560136192 }, { "epoch": 0.17, "learning_rate": 0.0004193380140421264, "loss": 2.909, "theoretical_loss": 3.8676732358919406, "tokens_seen": 560201728 }, { "epoch": 0.17, "learning_rate": 0.0004193279839518556, "loss": 2.7702, "theoretical_loss": 3.8676254555490788, "tokens_seen": 560267264 }, { "epoch": 0.17, "objective/train/docs_used": 327930, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8123157024383545, "objective/train/theoretical_loss": 3.8675776823595998, "objective/train/tokens_used": 580792800, "theoretical_loss": 3.8675776823595998, "tokens_seen": 560332800 }, { "epoch": 0.17, "learning_rate": 0.00041931795386158477, "loss": 2.6362, "theoretical_loss": 3.8675776823595998, "tokens_seen": 560332800 }, { "epoch": 0.17, "learning_rate": 0.00041930792377131395, "loss": 2.4958, "theoretical_loss": 3.867529916321595, "tokens_seen": 560398336 }, { "epoch": 0.17, "learning_rate": 0.00041929789368104313, "loss": 2.8404, "theoretical_loss": 3.867482157433157, "tokens_seen": 560463872 }, { "epoch": 0.17, "learning_rate": 0.0004192878635907723, "loss": 2.8702, "theoretical_loss": 3.8674344056923813, "tokens_seen": 560529408 }, { "epoch": 0.17, "learning_rate": 0.00041927783350050155, "loss": 2.9506, "theoretical_loss": 3.8673866610973624, "tokens_seen": 560594944 }, { "epoch": 0.17, "learning_rate": 0.00041926780341023067, "loss": 2.8685, "theoretical_loss": 3.8673389236461952, "tokens_seen": 560660480 }, { "epoch": 0.17, "learning_rate": 0.0004192577733199599, "loss": 2.8917, "theoretical_loss": 3.867291193336976, "tokens_seen": 560726016 }, { "epoch": 0.17, "learning_rate": 0.00041924774322968904, "loss": 2.8617, "theoretical_loss": 3.8672434701678022, "tokens_seen": 560791552 }, { "epoch": 0.17, "learning_rate": 0.00041923771313941827, "loss": 2.8296, "theoretical_loss": 3.867195754136771, "tokens_seen": 560857088 }, { "epoch": 0.17, "learning_rate": 0.00041922768304914745, "loss": 2.8242, "theoretical_loss": 3.867148045241981, "tokens_seen": 560922624 }, { "epoch": 0.17, "learning_rate": 0.00041921765295887663, "loss": 2.8429, "theoretical_loss": 3.867100343481532, "tokens_seen": 560988160 }, { "epoch": 0.17, "learning_rate": 0.0004192076228686058, "loss": 2.7912, "theoretical_loss": 3.8670526488535226, "tokens_seen": 561053696 }, { "epoch": 0.17, "learning_rate": 0.00041919759277833505, "loss": 2.696, "theoretical_loss": 3.867004961356054, "tokens_seen": 561119232 }, { "epoch": 0.17, "learning_rate": 0.0004191875626880642, "loss": 3.068, "theoretical_loss": 3.8669572809872284, "tokens_seen": 561184768 }, { "epoch": 0.17, "learning_rate": 0.0004191775325977934, "loss": 2.797, "theoretical_loss": 3.8669096077451464, "tokens_seen": 561250304 }, { "epoch": 0.17, "learning_rate": 0.00041916750250752254, "loss": 2.7436, "theoretical_loss": 3.8668619416279117, "tokens_seen": 561315840 }, { "epoch": 0.17, "learning_rate": 0.0004191574724172518, "loss": 2.8369, "theoretical_loss": 3.8668142826336274, "tokens_seen": 561381376 }, { "epoch": 0.17, "learning_rate": 0.00041914744232698095, "loss": 2.7877, "theoretical_loss": 3.866766630760398, "tokens_seen": 561446912 }, { "epoch": 0.17, "learning_rate": 0.00041913741223671014, "loss": 2.872, "theoretical_loss": 3.8667189860063282, "tokens_seen": 561512448 }, { "epoch": 0.17, "learning_rate": 0.0004191273821464393, "loss": 2.7135, "theoretical_loss": 3.866671348369524, "tokens_seen": 561577984 }, { "epoch": 0.17, "learning_rate": 0.0004191173520561685, "loss": 2.8627, "theoretical_loss": 3.866623717848092, "tokens_seen": 561643520 }, { "epoch": 0.17, "learning_rate": 0.0004191073219658977, "loss": 2.6897, "theoretical_loss": 3.8665760944401377, "tokens_seen": 561709056 }, { "epoch": 0.17, "learning_rate": 0.0004190972918756269, "loss": 3.0401, "theoretical_loss": 3.8665284781437714, "tokens_seen": 561774592 }, { "epoch": 0.17, "learning_rate": 0.00041908726178535604, "loss": 2.9005, "theoretical_loss": 3.8664808689571, "tokens_seen": 561840128 }, { "epoch": 0.17, "learning_rate": 0.0004190772316950853, "loss": 2.7574, "theoretical_loss": 3.866433266878234, "tokens_seen": 561905664 }, { "epoch": 0.17, "objective/train/docs_used": 328746, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.2547881603240967, "objective/train/theoretical_loss": 3.8663856719052827, "objective/train/tokens_used": 582431200, "theoretical_loss": 3.8663856719052827, "tokens_seen": 561971200 }, { "epoch": 0.17, "learning_rate": 0.0004190672016048144, "loss": 2.9363, "theoretical_loss": 3.8663856719052827, "tokens_seen": 561971200 }, { "epoch": 0.17, "learning_rate": 0.00041905717151454364, "loss": 3.0646, "theoretical_loss": 3.8663380840363564, "tokens_seen": 562036736 }, { "epoch": 0.17, "learning_rate": 0.0004190471414242728, "loss": 2.9747, "theoretical_loss": 3.866290503269567, "tokens_seen": 562102272 }, { "epoch": 0.17, "learning_rate": 0.000419037111334002, "loss": 2.915, "theoretical_loss": 3.866242929603027, "tokens_seen": 562167808 }, { "epoch": 0.17, "learning_rate": 0.0004190270812437312, "loss": 2.8532, "theoretical_loss": 3.8661953630348496, "tokens_seen": 562233344 }, { "epoch": 0.17, "learning_rate": 0.0004190170511534604, "loss": 2.8447, "theoretical_loss": 3.866147803563148, "tokens_seen": 562298880 }, { "epoch": 0.17, "learning_rate": 0.00041900702106318954, "loss": 3.027, "theoretical_loss": 3.866100251186036, "tokens_seen": 562364416 }, { "epoch": 0.17, "learning_rate": 0.0004189969909729188, "loss": 3.0567, "theoretical_loss": 3.8660527059016294, "tokens_seen": 562429952 }, { "epoch": 0.17, "learning_rate": 0.0004189869608826479, "loss": 2.8168, "theoretical_loss": 3.8660051677080434, "tokens_seen": 562495488 }, { "epoch": 0.17, "learning_rate": 0.00041897693079237714, "loss": 2.8508, "theoretical_loss": 3.8659576366033956, "tokens_seen": 562561024 }, { "epoch": 0.17, "learning_rate": 0.0004189669007021063, "loss": 2.6138, "theoretical_loss": 3.865910112585802, "tokens_seen": 562626560 }, { "epoch": 0.17, "learning_rate": 0.0004189568706118355, "loss": 2.9875, "theoretical_loss": 3.8658625956533816, "tokens_seen": 562692096 }, { "epoch": 0.17, "learning_rate": 0.0004189468405215647, "loss": 2.9688, "theoretical_loss": 3.8658150858042526, "tokens_seen": 562757632 }, { "epoch": 0.17, "learning_rate": 0.00041893681043129387, "loss": 3.0128, "theoretical_loss": 3.8657675830365346, "tokens_seen": 562823168 }, { "epoch": 0.17, "learning_rate": 0.00041892678034102305, "loss": 2.6135, "theoretical_loss": 3.865720087348347, "tokens_seen": 562888704 }, { "epoch": 0.17, "learning_rate": 0.0004189167502507523, "loss": 2.8912, "theoretical_loss": 3.8656725987378118, "tokens_seen": 562954240 }, { "epoch": 0.17, "learning_rate": 0.00041890672016048146, "loss": 2.9469, "theoretical_loss": 3.8656251172030496, "tokens_seen": 563019776 }, { "epoch": 0.17, "learning_rate": 0.00041889669007021065, "loss": 2.7757, "theoretical_loss": 3.865577642742183, "tokens_seen": 563085312 }, { "epoch": 0.17, "learning_rate": 0.0004188866599799399, "loss": 2.9993, "theoretical_loss": 3.865530175353335, "tokens_seen": 563150848 }, { "epoch": 0.17, "learning_rate": 0.000418876629889669, "loss": 2.7734, "theoretical_loss": 3.86548271503463, "tokens_seen": 563216384 }, { "epoch": 0.17, "learning_rate": 0.00041886659979939824, "loss": 2.6621, "theoretical_loss": 3.865435261784191, "tokens_seen": 563281920 }, { "epoch": 0.17, "learning_rate": 0.00041885656970912737, "loss": 3.0169, "theoretical_loss": 3.865387815600144, "tokens_seen": 563347456 }, { "epoch": 0.17, "learning_rate": 0.0004188465396188566, "loss": 3.0002, "theoretical_loss": 3.8653403764806145, "tokens_seen": 563412992 }, { "epoch": 0.17, "learning_rate": 0.0004188365095285858, "loss": 3.0746, "theoretical_loss": 3.86529294442373, "tokens_seen": 563478528 }, { "epoch": 0.17, "learning_rate": 0.00041882647943831497, "loss": 2.8611, "theoretical_loss": 3.8652455194276163, "tokens_seen": 563544064 }, { "epoch": 0.17, "objective/train/docs_used": 329425, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.942103862762451, "objective/train/theoretical_loss": 3.8651981014904027, "objective/train/tokens_used": 584069600, "theoretical_loss": 3.8651981014904027, "tokens_seen": 563609600 }, { "epoch": 0.17, "learning_rate": 0.00041881644934804415, "loss": 2.8762, "theoretical_loss": 3.8651981014904027, "tokens_seen": 563609600 }, { "epoch": 0.17, "learning_rate": 0.00041880641925777333, "loss": 2.8032, "theoretical_loss": 3.8651506906102173, "tokens_seen": 563675136 }, { "epoch": 0.17, "learning_rate": 0.0004187963891675025, "loss": 2.8483, "theoretical_loss": 3.8651032867851893, "tokens_seen": 563740672 }, { "epoch": 0.17, "learning_rate": 0.00041878635907723175, "loss": 3.0158, "theoretical_loss": 3.8650558900134495, "tokens_seen": 563806208 }, { "epoch": 0.17, "learning_rate": 0.0004187763289869609, "loss": 3.0126, "theoretical_loss": 3.8650085002931283, "tokens_seen": 563871744 }, { "epoch": 0.17, "learning_rate": 0.0004187662988966901, "loss": 3.0105, "theoretical_loss": 3.8649611176223573, "tokens_seen": 563937280 }, { "epoch": 0.17, "learning_rate": 0.00041875626880641924, "loss": 2.905, "theoretical_loss": 3.8649137419992687, "tokens_seen": 564002816 }, { "epoch": 0.17, "learning_rate": 0.00041874623871614847, "loss": 2.7749, "theoretical_loss": 3.864866373421996, "tokens_seen": 564068352 }, { "epoch": 0.17, "learning_rate": 0.00041873620862587765, "loss": 2.9877, "theoretical_loss": 3.8648190118886725, "tokens_seen": 564133888 }, { "epoch": 0.17, "learning_rate": 0.00041872617853560683, "loss": 2.8715, "theoretical_loss": 3.8647716573974327, "tokens_seen": 564199424 }, { "epoch": 0.17, "learning_rate": 0.000418716148445336, "loss": 2.7122, "theoretical_loss": 3.8647243099464106, "tokens_seen": 564264960 }, { "epoch": 0.17, "learning_rate": 0.00041870611835506525, "loss": 2.9309, "theoretical_loss": 3.864676969533744, "tokens_seen": 564330496 }, { "epoch": 0.17, "learning_rate": 0.0004186960882647944, "loss": 2.7819, "theoretical_loss": 3.8646296361575687, "tokens_seen": 564396032 }, { "epoch": 0.17, "learning_rate": 0.0004186860581745236, "loss": 2.7789, "theoretical_loss": 3.8645823098160212, "tokens_seen": 564461568 }, { "epoch": 0.17, "learning_rate": 0.00041867602808425274, "loss": 3.1001, "theoretical_loss": 3.8645349905072406, "tokens_seen": 564527104 }, { "epoch": 0.17, "learning_rate": 0.000418665997993982, "loss": 2.7691, "theoretical_loss": 3.8644876782293647, "tokens_seen": 564592640 }, { "epoch": 0.17, "learning_rate": 0.00041865596790371115, "loss": 3.0858, "theoretical_loss": 3.864440372980533, "tokens_seen": 564658176 }, { "epoch": 0.17, "learning_rate": 0.00041864593781344034, "loss": 2.9535, "theoretical_loss": 3.8643930747588855, "tokens_seen": 564723712 }, { "epoch": 0.17, "learning_rate": 0.0004186359077231695, "loss": 2.5333, "theoretical_loss": 3.8643457835625634, "tokens_seen": 564789248 }, { "epoch": 0.17, "learning_rate": 0.0004186258776328987, "loss": 2.9543, "theoretical_loss": 3.8642984993897085, "tokens_seen": 564854784 }, { "epoch": 0.17, "learning_rate": 0.0004186158475426279, "loss": 2.5846, "theoretical_loss": 3.864251222238462, "tokens_seen": 564920320 }, { "epoch": 0.17, "learning_rate": 0.0004186058174523571, "loss": 3.0025, "theoretical_loss": 3.8642039521069678, "tokens_seen": 564985856 }, { "epoch": 0.17, "learning_rate": 0.00041859578736208624, "loss": 2.9297, "theoretical_loss": 3.864156688993369, "tokens_seen": 565051392 }, { "epoch": 0.17, "learning_rate": 0.0004185857572718155, "loss": 2.944, "theoretical_loss": 3.86410943289581, "tokens_seen": 565116928 }, { "epoch": 0.17, "learning_rate": 0.0004185757271815446, "loss": 2.6911, "theoretical_loss": 3.864062183812436, "tokens_seen": 565182464 }, { "epoch": 0.17, "objective/train/docs_used": 329612, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.730563163757324, "objective/train/theoretical_loss": 3.864014941741392, "objective/train/tokens_used": 585708000, "theoretical_loss": 3.864014941741392, "tokens_seen": 565248000 }, { "epoch": 0.17, "learning_rate": 0.00041856569709127384, "loss": 3.1174, "theoretical_loss": 3.864014941741392, "tokens_seen": 565248000 }, { "epoch": 0.17, "learning_rate": 0.000418555667001003, "loss": 3.1198, "theoretical_loss": 3.8639677066808256, "tokens_seen": 565313536 }, { "epoch": 0.17, "learning_rate": 0.0004185456369107322, "loss": 2.9859, "theoretical_loss": 3.863920478628883, "tokens_seen": 565379072 }, { "epoch": 0.17, "learning_rate": 0.0004185356068204614, "loss": 2.9614, "theoretical_loss": 3.8638732575837125, "tokens_seen": 565444608 }, { "epoch": 0.17, "learning_rate": 0.0004185255767301906, "loss": 2.7855, "theoretical_loss": 3.8638260435434626, "tokens_seen": 565510144 }, { "epoch": 0.17, "learning_rate": 0.00041851554663991974, "loss": 2.9431, "theoretical_loss": 3.863778836506283, "tokens_seen": 565575680 }, { "epoch": 0.17, "learning_rate": 0.000418505516549649, "loss": 2.7765, "theoretical_loss": 3.8637316364703227, "tokens_seen": 565641216 }, { "epoch": 0.17, "learning_rate": 0.0004184954864593781, "loss": 2.9082, "theoretical_loss": 3.8636844434337325, "tokens_seen": 565706752 }, { "epoch": 0.17, "learning_rate": 0.00041848545636910734, "loss": 3.0147, "theoretical_loss": 3.863637257394665, "tokens_seen": 565772288 }, { "epoch": 0.17, "learning_rate": 0.0004184754262788365, "loss": 3.0763, "theoretical_loss": 3.863590078351271, "tokens_seen": 565837824 }, { "epoch": 0.17, "learning_rate": 0.0004184653961885657, "loss": 2.9114, "theoretical_loss": 3.8635429063017037, "tokens_seen": 565903360 }, { "epoch": 0.17, "learning_rate": 0.0004184553660982949, "loss": 2.7337, "theoretical_loss": 3.8634957412441167, "tokens_seen": 565968896 }, { "epoch": 0.17, "learning_rate": 0.00041844533600802407, "loss": 2.9751, "theoretical_loss": 3.8634485831766634, "tokens_seen": 566034432 }, { "epoch": 0.17, "learning_rate": 0.00041843530591775325, "loss": 2.9623, "theoretical_loss": 3.8634014320975005, "tokens_seen": 566099968 }, { "epoch": 0.17, "learning_rate": 0.0004184252758274825, "loss": 3.0314, "theoretical_loss": 3.8633542880047815, "tokens_seen": 566165504 }, { "epoch": 0.17, "learning_rate": 0.0004184152457372116, "loss": 2.9415, "theoretical_loss": 3.863307150896664, "tokens_seen": 566231040 }, { "epoch": 0.17, "learning_rate": 0.00041840521564694085, "loss": 2.6393, "theoretical_loss": 3.863260020771304, "tokens_seen": 566296576 }, { "epoch": 0.17, "learning_rate": 0.00041839518555666997, "loss": 3.0198, "theoretical_loss": 3.8632128976268603, "tokens_seen": 566362112 }, { "epoch": 0.17, "learning_rate": 0.0004183851554663992, "loss": 2.9247, "theoretical_loss": 3.8631657814614906, "tokens_seen": 566427648 }, { "epoch": 0.17, "learning_rate": 0.0004183751253761284, "loss": 3.0689, "theoretical_loss": 3.8631186722733535, "tokens_seen": 566493184 }, { "epoch": 0.17, "learning_rate": 0.00041836509528585757, "loss": 2.8912, "theoretical_loss": 3.86307157006061, "tokens_seen": 566558720 }, { "epoch": 0.17, "learning_rate": 0.00041835506519558675, "loss": 2.8435, "theoretical_loss": 3.8630244748214198, "tokens_seen": 566624256 }, { "epoch": 0.17, "learning_rate": 0.000418345035105316, "loss": 2.9153, "theoretical_loss": 3.862977386553944, "tokens_seen": 566689792 }, { "epoch": 0.17, "learning_rate": 0.0004183350050150451, "loss": 2.7937, "theoretical_loss": 3.862930305256345, "tokens_seen": 566755328 }, { "epoch": 0.17, "learning_rate": 0.00041832497492477435, "loss": 3.0194, "theoretical_loss": 3.8628832309267844, "tokens_seen": 566820864 }, { "epoch": 0.17, "objective/train/docs_used": 331071, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8123180866241455, "objective/train/theoretical_loss": 3.8628361635634265, "objective/train/tokens_used": 587346400, "theoretical_loss": 3.8628361635634265, "tokens_seen": 566886400 }, { "epoch": 0.17, "learning_rate": 0.0004183149448345035, "loss": 2.6294, "theoretical_loss": 3.8628361635634265, "tokens_seen": 566886400 }, { "epoch": 0.17, "learning_rate": 0.0004183049147442327, "loss": 3.1253, "theoretical_loss": 3.862789103164435, "tokens_seen": 566951936 }, { "epoch": 0.17, "learning_rate": 0.0004182948846539619, "loss": 2.8894, "theoretical_loss": 3.8627420497279745, "tokens_seen": 567017472 }, { "epoch": 0.17, "learning_rate": 0.0004182848545636911, "loss": 2.9479, "theoretical_loss": 3.8626950032522096, "tokens_seen": 567083008 }, { "epoch": 0.17, "learning_rate": 0.00041827482447342025, "loss": 2.9446, "theoretical_loss": 3.8626479637353075, "tokens_seen": 567148544 }, { "epoch": 0.17, "learning_rate": 0.00041826479438314944, "loss": 2.9645, "theoretical_loss": 3.862600931175434, "tokens_seen": 567214080 }, { "epoch": 0.17, "learning_rate": 0.0004182547642928786, "loss": 2.9893, "theoretical_loss": 3.8625539055707576, "tokens_seen": 567279616 }, { "epoch": 0.17, "learning_rate": 0.00041824473420260785, "loss": 2.9431, "theoretical_loss": 3.862506886919445, "tokens_seen": 567345152 }, { "epoch": 0.17, "learning_rate": 0.000418234704112337, "loss": 2.8937, "theoretical_loss": 3.862459875219667, "tokens_seen": 567410688 }, { "epoch": 0.17, "learning_rate": 0.0004182246740220662, "loss": 3.0324, "theoretical_loss": 3.862412870469591, "tokens_seen": 567476224 }, { "epoch": 0.17, "learning_rate": 0.0004182146439317954, "loss": 2.989, "theoretical_loss": 3.8623658726673886, "tokens_seen": 567541760 }, { "epoch": 0.17, "learning_rate": 0.0004182046138415246, "loss": 3.095, "theoretical_loss": 3.86231888181123, "tokens_seen": 567607296 }, { "epoch": 0.17, "learning_rate": 0.00041819458375125376, "loss": 2.8723, "theoretical_loss": 3.862271897899287, "tokens_seen": 567672832 }, { "epoch": 0.17, "learning_rate": 0.00041818455366098294, "loss": 2.847, "theoretical_loss": 3.862224920929732, "tokens_seen": 567738368 }, { "epoch": 0.17, "learning_rate": 0.0004181745235707121, "loss": 2.9045, "theoretical_loss": 3.8621779509007377, "tokens_seen": 567803904 }, { "epoch": 0.17, "learning_rate": 0.00041816449348044136, "loss": 2.9238, "theoretical_loss": 3.8621309878104784, "tokens_seen": 567869440 }, { "epoch": 0.17, "learning_rate": 0.00041815446339017054, "loss": 2.8785, "theoretical_loss": 3.862084031657128, "tokens_seen": 567934976 }, { "epoch": 0.17, "learning_rate": 0.0004181444332998997, "loss": 2.7955, "theoretical_loss": 3.862037082438861, "tokens_seen": 568000512 }, { "epoch": 0.17, "learning_rate": 0.0004181344032096289, "loss": 2.9675, "theoretical_loss": 3.8619901401538543, "tokens_seen": 568066048 }, { "epoch": 0.17, "learning_rate": 0.0004181243731193581, "loss": 2.691, "theoretical_loss": 3.8619432048002835, "tokens_seen": 568131584 }, { "epoch": 0.17, "learning_rate": 0.0004181143430290873, "loss": 3.0101, "theoretical_loss": 3.861896276376326, "tokens_seen": 568197120 }, { "epoch": 0.17, "learning_rate": 0.00041810431293881644, "loss": 2.9619, "theoretical_loss": 3.8618493548801602, "tokens_seen": 568262656 }, { "epoch": 0.17, "learning_rate": 0.0004180942828485457, "loss": 2.9554, "theoretical_loss": 3.8618024403099636, "tokens_seen": 568328192 }, { "epoch": 0.17, "learning_rate": 0.0004180842527582748, "loss": 3.0257, "theoretical_loss": 3.8617555326639157, "tokens_seen": 568393728 }, { "epoch": 0.17, "learning_rate": 0.00041807422266800404, "loss": 3.0616, "theoretical_loss": 3.8617086319401968, "tokens_seen": 568459264 }, { "epoch": 0.17, "objective/train/docs_used": 331416, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.896811008453369, "objective/train/theoretical_loss": 3.861661738136987, "objective/train/tokens_used": 588984800, "theoretical_loss": 3.861661738136987, "tokens_seen": 568524800 }, { "epoch": 0.17, "learning_rate": 0.0004180641925777332, "loss": 2.9509, "theoretical_loss": 3.861661738136987, "tokens_seen": 568524800 }, { "epoch": 0.17, "learning_rate": 0.0004180541624874624, "loss": 2.8699, "theoretical_loss": 3.8616148512524675, "tokens_seen": 568590336 }, { "epoch": 0.17, "learning_rate": 0.0004180441323971916, "loss": 2.9458, "theoretical_loss": 3.8615679712848205, "tokens_seen": 568655872 }, { "epoch": 0.17, "learning_rate": 0.0004180341023069208, "loss": 3.1109, "theoretical_loss": 3.8615210982322288, "tokens_seen": 568721408 }, { "epoch": 0.17, "learning_rate": 0.00041802407221664994, "loss": 3.0128, "theoretical_loss": 3.861474232092876, "tokens_seen": 568786944 }, { "epoch": 0.17, "learning_rate": 0.0004180140421263792, "loss": 3.1939, "theoretical_loss": 3.8614273728649446, "tokens_seen": 568852480 }, { "epoch": 0.17, "learning_rate": 0.0004180040120361083, "loss": 2.831, "theoretical_loss": 3.8613805205466205, "tokens_seen": 568918016 }, { "epoch": 0.17, "learning_rate": 0.00041799398194583754, "loss": 2.7662, "theoretical_loss": 3.8613336751360894, "tokens_seen": 568983552 }, { "epoch": 0.17, "learning_rate": 0.0004179839518555667, "loss": 2.8514, "theoretical_loss": 3.861286836631537, "tokens_seen": 569049088 }, { "epoch": 0.17, "learning_rate": 0.0004179739217652959, "loss": 2.6863, "theoretical_loss": 3.8612400050311493, "tokens_seen": 569114624 }, { "epoch": 0.17, "learning_rate": 0.0004179638916750251, "loss": 2.9529, "theoretical_loss": 3.8611931803331148, "tokens_seen": 569180160 }, { "epoch": 0.17, "learning_rate": 0.00041795386158475427, "loss": 2.7516, "theoretical_loss": 3.861146362535621, "tokens_seen": 569245696 }, { "epoch": 0.17, "learning_rate": 0.00041794383149448345, "loss": 3.1024, "theoretical_loss": 3.8610995516368565, "tokens_seen": 569311232 }, { "epoch": 0.17, "learning_rate": 0.0004179338014042127, "loss": 2.8311, "theoretical_loss": 3.8610527476350116, "tokens_seen": 569376768 }, { "epoch": 0.17, "learning_rate": 0.0004179237713139418, "loss": 2.7588, "theoretical_loss": 3.8610059505282757, "tokens_seen": 569442304 }, { "epoch": 0.17, "learning_rate": 0.00041791374122367105, "loss": 2.7985, "theoretical_loss": 3.86095916031484, "tokens_seen": 569507840 }, { "epoch": 0.17, "learning_rate": 0.00041790371113340017, "loss": 2.8401, "theoretical_loss": 3.860912376992896, "tokens_seen": 569573376 }, { "epoch": 0.17, "learning_rate": 0.0004178936810431294, "loss": 2.9647, "theoretical_loss": 3.8608656005606354, "tokens_seen": 569638912 }, { "epoch": 0.17, "learning_rate": 0.0004178836509528586, "loss": 2.8687, "theoretical_loss": 3.8608188310162523, "tokens_seen": 569704448 }, { "epoch": 0.17, "learning_rate": 0.00041787362086258777, "loss": 2.6109, "theoretical_loss": 3.8607720683579396, "tokens_seen": 569769984 }, { "epoch": 0.17, "learning_rate": 0.00041786359077231695, "loss": 3.1195, "theoretical_loss": 3.8607253125838916, "tokens_seen": 569835520 }, { "epoch": 0.17, "learning_rate": 0.0004178535606820462, "loss": 2.9842, "theoretical_loss": 3.8606785636923027, "tokens_seen": 569901056 }, { "epoch": 0.17, "learning_rate": 0.0004178435305917753, "loss": 2.754, "theoretical_loss": 3.8606318216813693, "tokens_seen": 569966592 }, { "epoch": 0.17, "learning_rate": 0.00041783350050150455, "loss": 2.779, "theoretical_loss": 3.8605850865492872, "tokens_seen": 570032128 }, { "epoch": 0.17, "learning_rate": 0.0004178234704112337, "loss": 3.0349, "theoretical_loss": 3.860538358294254, "tokens_seen": 570097664 }, { "epoch": 0.17, "objective/train/docs_used": 332819, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.92439341545105, "objective/train/theoretical_loss": 3.8604916369144666, "objective/train/tokens_used": 590623200, "theoretical_loss": 3.8604916369144666, "tokens_seen": 570163200 }, { "epoch": 0.17, "learning_rate": 0.0004178134403209629, "loss": 3.011, "theoretical_loss": 3.8604916369144666, "tokens_seen": 570163200 }, { "epoch": 0.17, "learning_rate": 0.0004178034102306921, "loss": 2.9937, "theoretical_loss": 3.8604449224081234, "tokens_seen": 570228736 }, { "epoch": 0.17, "learning_rate": 0.0004177933801404213, "loss": 2.959, "theoretical_loss": 3.860398214773425, "tokens_seen": 570294272 }, { "epoch": 0.17, "learning_rate": 0.00041778335005015045, "loss": 3.0047, "theoretical_loss": 3.8603515140085687, "tokens_seen": 570359808 }, { "epoch": 0.17, "learning_rate": 0.00041777331995987964, "loss": 2.81, "theoretical_loss": 3.8603048201117565, "tokens_seen": 570425344 }, { "epoch": 0.17, "learning_rate": 0.0004177632898696088, "loss": 2.8243, "theoretical_loss": 3.8602581330811887, "tokens_seen": 570490880 }, { "epoch": 0.17, "learning_rate": 0.00041775325977933805, "loss": 2.9167, "theoretical_loss": 3.8602114529150673, "tokens_seen": 570556416 }, { "epoch": 0.17, "learning_rate": 0.0004177432296890672, "loss": 2.8378, "theoretical_loss": 3.860164779611595, "tokens_seen": 570621952 }, { "epoch": 0.17, "learning_rate": 0.0004177331995987964, "loss": 2.6981, "theoretical_loss": 3.8601181131689737, "tokens_seen": 570687488 }, { "epoch": 0.17, "learning_rate": 0.0004177231695085256, "loss": 2.8103, "theoretical_loss": 3.8600714535854093, "tokens_seen": 570753024 }, { "epoch": 0.17, "learning_rate": 0.0004177131394182548, "loss": 2.8283, "theoretical_loss": 3.8600248008591045, "tokens_seen": 570818560 }, { "epoch": 0.17, "learning_rate": 0.00041770310932798396, "loss": 2.7991, "theoretical_loss": 3.859978154988265, "tokens_seen": 570884096 }, { "epoch": 0.17, "learning_rate": 0.00041769307923771314, "loss": 2.8655, "theoretical_loss": 3.8599315159710965, "tokens_seen": 570949632 }, { "epoch": 0.17, "learning_rate": 0.0004176830491474423, "loss": 2.8578, "theoretical_loss": 3.8598848838058055, "tokens_seen": 571015168 }, { "epoch": 0.17, "learning_rate": 0.00041767301905717156, "loss": 2.8153, "theoretical_loss": 3.8598382584905995, "tokens_seen": 571080704 }, { "epoch": 0.17, "learning_rate": 0.0004176629889669007, "loss": 3.1572, "theoretical_loss": 3.859791640023685, "tokens_seen": 571146240 }, { "epoch": 0.17, "learning_rate": 0.0004176529588766299, "loss": 2.7899, "theoretical_loss": 3.859745028403273, "tokens_seen": 571211776 }, { "epoch": 0.17, "learning_rate": 0.00041764292878635904, "loss": 2.7977, "theoretical_loss": 3.85969842362757, "tokens_seen": 571277312 }, { "epoch": 0.17, "learning_rate": 0.0004176328986960883, "loss": 2.98, "theoretical_loss": 3.859651825694787, "tokens_seen": 571342848 }, { "epoch": 0.17, "learning_rate": 0.00041762286860581746, "loss": 2.8098, "theoretical_loss": 3.859605234603135, "tokens_seen": 571408384 }, { "epoch": 0.17, "learning_rate": 0.00041761283851554664, "loss": 2.8236, "theoretical_loss": 3.8595586503508246, "tokens_seen": 571473920 }, { "epoch": 0.17, "learning_rate": 0.0004176028084252758, "loss": 3.083, "theoretical_loss": 3.8595120729360675, "tokens_seen": 571539456 }, { "epoch": 0.17, "learning_rate": 0.000417592778335005, "loss": 2.711, "theoretical_loss": 3.8594655023570765, "tokens_seen": 571604992 }, { "epoch": 0.17, "learning_rate": 0.0004175827482447342, "loss": 2.8807, "theoretical_loss": 3.859418938612065, "tokens_seen": 571670528 }, { "epoch": 0.17, "learning_rate": 0.0004175727181544634, "loss": 3.1158, "theoretical_loss": 3.859372381699247, "tokens_seen": 571736064 }, { "epoch": 0.17, "objective/train/docs_used": 333556, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.679062604904175, "objective/train/theoretical_loss": 3.859325831616836, "objective/train/tokens_used": 592261600, "theoretical_loss": 3.859325831616836, "tokens_seen": 571801600 }, { "epoch": 0.17, "learning_rate": 0.00041756268806419255, "loss": 2.6757, "theoretical_loss": 3.859325831616836, "tokens_seen": 571801600 }, { "epoch": 0.17, "learning_rate": 0.0004175526579739218, "loss": 2.8884, "theoretical_loss": 3.859279288363048, "tokens_seen": 571867136 }, { "epoch": 0.17, "learning_rate": 0.00041754262788365096, "loss": 3.0039, "theoretical_loss": 3.8592327519360996, "tokens_seen": 571932672 }, { "epoch": 0.17, "learning_rate": 0.00041753259779338015, "loss": 2.6379, "theoretical_loss": 3.8591862223342064, "tokens_seen": 571998208 }, { "epoch": 0.17, "learning_rate": 0.0004175225677031093, "loss": 2.8799, "theoretical_loss": 3.8591396995555853, "tokens_seen": 572063744 }, { "epoch": 0.17, "learning_rate": 0.0004175125376128385, "loss": 2.9484, "theoretical_loss": 3.859093183598455, "tokens_seen": 572129280 }, { "epoch": 0.17, "learning_rate": 0.0004175025075225677, "loss": 2.8176, "theoretical_loss": 3.859046674461034, "tokens_seen": 572194816 }, { "epoch": 0.17, "learning_rate": 0.0004174924774322969, "loss": 2.8921, "theoretical_loss": 3.859000172141541, "tokens_seen": 572260352 }, { "epoch": 0.17, "learning_rate": 0.00041748244734202605, "loss": 2.8583, "theoretical_loss": 3.8589536766381967, "tokens_seen": 572325888 }, { "epoch": 0.17, "learning_rate": 0.0004174724172517553, "loss": 2.9911, "theoretical_loss": 3.858907187949221, "tokens_seen": 572391424 }, { "epoch": 0.17, "learning_rate": 0.0004174623871614844, "loss": 2.6848, "theoretical_loss": 3.8588607060728357, "tokens_seen": 572456960 }, { "epoch": 0.17, "learning_rate": 0.00041745235707121365, "loss": 2.8667, "theoretical_loss": 3.858814231007262, "tokens_seen": 572522496 }, { "epoch": 0.17, "learning_rate": 0.00041744232698094283, "loss": 3.1139, "theoretical_loss": 3.8587677627507233, "tokens_seen": 572588032 }, { "epoch": 0.17, "learning_rate": 0.000417432296890672, "loss": 2.8625, "theoretical_loss": 3.858721301301442, "tokens_seen": 572653568 }, { "epoch": 0.17, "learning_rate": 0.0004174222668004012, "loss": 2.9168, "theoretical_loss": 3.8586748466576424, "tokens_seen": 572719104 }, { "epoch": 0.17, "learning_rate": 0.00041741223671013037, "loss": 3.0097, "theoretical_loss": 3.8586283988175496, "tokens_seen": 572784640 }, { "epoch": 0.17, "learning_rate": 0.0004174022066198596, "loss": 2.9648, "theoretical_loss": 3.8585819577793883, "tokens_seen": 572850176 }, { "epoch": 0.17, "learning_rate": 0.0004173921765295888, "loss": 2.7055, "theoretical_loss": 3.8585355235413843, "tokens_seen": 572915712 }, { "epoch": 0.17, "learning_rate": 0.00041738214643931797, "loss": 2.6896, "theoretical_loss": 3.858489096101765, "tokens_seen": 572981248 }, { "epoch": 0.17, "learning_rate": 0.00041737211634904715, "loss": 2.8474, "theoretical_loss": 3.858442675458756, "tokens_seen": 573046784 }, { "epoch": 0.17, "learning_rate": 0.0004173620862587764, "loss": 2.8638, "theoretical_loss": 3.8583962616105874, "tokens_seen": 573112320 }, { "epoch": 0.17, "learning_rate": 0.0004173520561685055, "loss": 3.0345, "theoretical_loss": 3.858349854555486, "tokens_seen": 573177856 }, { "epoch": 0.17, "learning_rate": 0.00041734202607823475, "loss": 2.8815, "theoretical_loss": 3.858303454291682, "tokens_seen": 573243392 }, { "epoch": 0.17, "learning_rate": 0.0004173319959879639, "loss": 2.9809, "theoretical_loss": 3.8582570608174045, "tokens_seen": 573308928 }, { "epoch": 0.17, "learning_rate": 0.0004173219658976931, "loss": 2.8129, "theoretical_loss": 3.858210674130885, "tokens_seen": 573374464 }, { "epoch": 0.17, "objective/train/docs_used": 334721, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6601226329803467, "objective/train/theoretical_loss": 3.858164294230354, "objective/train/tokens_used": 593900000, "theoretical_loss": 3.858164294230354, "tokens_seen": 573440000 }, { "epoch": 0.17, "learning_rate": 0.0004173119358074223, "loss": 3.0888, "theoretical_loss": 3.858164294230354, "tokens_seen": 573440000 }, { "epoch": 0.17, "learning_rate": 0.0004173019057171515, "loss": 2.9326, "theoretical_loss": 3.8581179211140437, "tokens_seen": 573505536 }, { "epoch": 0.17, "learning_rate": 0.00041729187562688065, "loss": 2.9435, "theoretical_loss": 3.8580715547801865, "tokens_seen": 573571072 }, { "epoch": 0.17, "learning_rate": 0.00041728184553660984, "loss": 2.777, "theoretical_loss": 3.8580251952270164, "tokens_seen": 573636608 }, { "epoch": 0.17, "learning_rate": 0.000417271815446339, "loss": 2.9294, "theoretical_loss": 3.8579788424527655, "tokens_seen": 573702144 }, { "epoch": 0.17, "learning_rate": 0.00041726178535606825, "loss": 3.1094, "theoretical_loss": 3.8579324964556703, "tokens_seen": 573767680 }, { "epoch": 0.17, "learning_rate": 0.0004172517552657974, "loss": 3.2043, "theoretical_loss": 3.857886157233965, "tokens_seen": 573833216 }, { "epoch": 0.17, "learning_rate": 0.0004172417251755266, "loss": 2.8649, "theoretical_loss": 3.8578398247858856, "tokens_seen": 573898752 }, { "epoch": 0.17, "learning_rate": 0.0004172316950852558, "loss": 2.6749, "theoretical_loss": 3.857793499109669, "tokens_seen": 573964288 }, { "epoch": 0.17, "learning_rate": 0.000417221664994985, "loss": 2.7361, "theoretical_loss": 3.8577471802035515, "tokens_seen": 574029824 }, { "epoch": 0.17, "learning_rate": 0.00041721163490471416, "loss": 2.934, "theoretical_loss": 3.8577008680657716, "tokens_seen": 574095360 }, { "epoch": 0.17, "learning_rate": 0.00041720160481444334, "loss": 2.8451, "theoretical_loss": 3.8576545626945684, "tokens_seen": 574160896 }, { "epoch": 0.17, "learning_rate": 0.0004171915747241725, "loss": 2.7165, "theoretical_loss": 3.8576082640881797, "tokens_seen": 574226432 }, { "epoch": 0.17, "learning_rate": 0.00041718154463390176, "loss": 2.7004, "theoretical_loss": 3.857561972244846, "tokens_seen": 574291968 }, { "epoch": 0.17, "learning_rate": 0.0004171715145436309, "loss": 2.9106, "theoretical_loss": 3.8575156871628082, "tokens_seen": 574357504 }, { "epoch": 0.17, "learning_rate": 0.0004171614844533601, "loss": 2.984, "theoretical_loss": 3.8574694088403065, "tokens_seen": 574423040 }, { "epoch": 0.17, "learning_rate": 0.00041715145436308924, "loss": 2.8751, "theoretical_loss": 3.8574231372755836, "tokens_seen": 574488576 }, { "epoch": 0.17, "learning_rate": 0.0004171414242728185, "loss": 2.7598, "theoretical_loss": 3.8573768724668813, "tokens_seen": 574554112 }, { "epoch": 0.17, "learning_rate": 0.00041713139418254766, "loss": 2.8516, "theoretical_loss": 3.8573306144124437, "tokens_seen": 574619648 }, { "epoch": 0.17, "learning_rate": 0.00041712136409227684, "loss": 2.8133, "theoretical_loss": 3.857284363110513, "tokens_seen": 574685184 }, { "epoch": 0.17, "learning_rate": 0.000417111334002006, "loss": 2.7077, "theoretical_loss": 3.8572381185593354, "tokens_seen": 574750720 }, { "epoch": 0.17, "learning_rate": 0.0004171013039117352, "loss": 3.0123, "theoretical_loss": 3.8571918807571546, "tokens_seen": 574816256 }, { "epoch": 0.17, "learning_rate": 0.0004170912738214644, "loss": 2.8599, "theoretical_loss": 3.8571456497022165, "tokens_seen": 574881792 }, { "epoch": 0.17, "learning_rate": 0.0004170812437311936, "loss": 2.9078, "theoretical_loss": 3.857099425392768, "tokens_seen": 574947328 }, { "epoch": 0.17, "learning_rate": 0.00041707121364092275, "loss": 2.8265, "theoretical_loss": 3.8570532078270565, "tokens_seen": 575012864 }, { "epoch": 0.17, "objective/train/docs_used": 335233, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9721813201904297, "objective/train/theoretical_loss": 3.857006997003329, "objective/train/tokens_used": 595538400, "theoretical_loss": 3.857006997003329, "tokens_seen": 575078400 }, { "epoch": 0.17, "learning_rate": 0.000417061183550652, "loss": 2.9865, "theoretical_loss": 3.857006997003329, "tokens_seen": 575078400 }, { "epoch": 0.17, "learning_rate": 0.00041705115346038116, "loss": 2.6959, "theoretical_loss": 3.8569607929198337, "tokens_seen": 575143936 }, { "epoch": 0.17, "learning_rate": 0.00041704112337011035, "loss": 2.9934, "theoretical_loss": 3.85691459557482, "tokens_seen": 575209472 }, { "epoch": 0.17, "learning_rate": 0.0004170310932798395, "loss": 2.6767, "theoretical_loss": 3.856868404966538, "tokens_seen": 575275008 }, { "epoch": 0.17, "learning_rate": 0.0004170210631895687, "loss": 2.934, "theoretical_loss": 3.8568222210932372, "tokens_seen": 575340544 }, { "epoch": 0.17, "learning_rate": 0.0004170110330992979, "loss": 2.8967, "theoretical_loss": 3.856776043953169, "tokens_seen": 575406080 }, { "epoch": 0.17, "learning_rate": 0.0004170010030090271, "loss": 2.8635, "theoretical_loss": 3.8567298735445847, "tokens_seen": 575471616 }, { "epoch": 0.17, "learning_rate": 0.00041699097291875625, "loss": 2.6781, "theoretical_loss": 3.856683709865737, "tokens_seen": 575537152 }, { "epoch": 0.17, "learning_rate": 0.0004169809428284855, "loss": 2.9516, "theoretical_loss": 3.8566375529148793, "tokens_seen": 575602688 }, { "epoch": 0.17, "learning_rate": 0.0004169709127382146, "loss": 2.7865, "theoretical_loss": 3.856591402690264, "tokens_seen": 575668224 }, { "epoch": 0.17, "learning_rate": 0.00041696088264794385, "loss": 2.7155, "theoretical_loss": 3.8565452591901463, "tokens_seen": 575733760 }, { "epoch": 0.17, "learning_rate": 0.00041695085255767303, "loss": 2.892, "theoretical_loss": 3.856499122412781, "tokens_seen": 575799296 }, { "epoch": 0.17, "learning_rate": 0.0004169408224674022, "loss": 3.0308, "theoretical_loss": 3.8564529923564232, "tokens_seen": 575864832 }, { "epoch": 0.17, "learning_rate": 0.0004169307923771314, "loss": 2.7946, "theoretical_loss": 3.856406869019329, "tokens_seen": 575930368 }, { "epoch": 0.17, "learning_rate": 0.0004169207622868606, "loss": 2.8613, "theoretical_loss": 3.8563607523997563, "tokens_seen": 575995904 }, { "epoch": 0.17, "learning_rate": 0.00041691073219658975, "loss": 2.8431, "theoretical_loss": 3.856314642495961, "tokens_seen": 576061440 }, { "epoch": 0.17, "learning_rate": 0.000416900702106319, "loss": 2.8555, "theoretical_loss": 3.8562685393062024, "tokens_seen": 576126976 }, { "epoch": 0.17, "learning_rate": 0.0004168906720160481, "loss": 2.857, "theoretical_loss": 3.8562224428287393, "tokens_seen": 576192512 }, { "epoch": 0.17, "learning_rate": 0.00041688064192577735, "loss": 2.8366, "theoretical_loss": 3.85617635306183, "tokens_seen": 576258048 }, { "epoch": 0.17, "learning_rate": 0.00041687061183550653, "loss": 2.831, "theoretical_loss": 3.856130270003736, "tokens_seen": 576323584 }, { "epoch": 0.17, "learning_rate": 0.0004168605817452357, "loss": 2.8643, "theoretical_loss": 3.856084193652718, "tokens_seen": 576389120 }, { "epoch": 0.17, "learning_rate": 0.0004168505516549649, "loss": 2.6865, "theoretical_loss": 3.8560381240070365, "tokens_seen": 576454656 }, { "epoch": 0.17, "learning_rate": 0.0004168405215646941, "loss": 2.8393, "theoretical_loss": 3.8559920610649536, "tokens_seen": 576520192 }, { "epoch": 0.17, "learning_rate": 0.00041683049147442326, "loss": 2.7828, "theoretical_loss": 3.855946004824733, "tokens_seen": 576585728 }, { "epoch": 0.17, "learning_rate": 0.0004168204613841525, "loss": 2.7994, "theoretical_loss": 3.8558999552846367, "tokens_seen": 576651264 }, { "epoch": 0.17, "objective/train/docs_used": 335838, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.021301031112671, "objective/train/theoretical_loss": 3.85585391244293, "objective/train/tokens_used": 597176800, "theoretical_loss": 3.85585391244293, "tokens_seen": 576716800 }, { "epoch": 0.17, "learning_rate": 0.0004168104312938816, "loss": 2.7419, "theoretical_loss": 3.85585391244293, "tokens_seen": 576716800 }, { "epoch": 0.17, "learning_rate": 0.00041680040120361085, "loss": 2.8041, "theoretical_loss": 3.855807876297876, "tokens_seen": 576782336 }, { "epoch": 0.17, "learning_rate": 0.00041679037111334, "loss": 2.6524, "theoretical_loss": 3.8557618468477415, "tokens_seen": 576847872 }, { "epoch": 0.17, "learning_rate": 0.0004167803410230692, "loss": 3.0374, "theoretical_loss": 3.8557158240907916, "tokens_seen": 576913408 }, { "epoch": 0.17, "learning_rate": 0.0004167703109327984, "loss": 3.1909, "theoretical_loss": 3.8556698080252936, "tokens_seen": 576978944 }, { "epoch": 0.17, "learning_rate": 0.0004167602808425276, "loss": 2.6784, "theoretical_loss": 3.8556237986495137, "tokens_seen": 577044480 }, { "epoch": 0.17, "learning_rate": 0.00041675025075225676, "loss": 2.6979, "theoretical_loss": 3.8555777959617203, "tokens_seen": 577110016 }, { "epoch": 0.17, "learning_rate": 0.000416740220661986, "loss": 2.9514, "theoretical_loss": 3.8555317999601817, "tokens_seen": 577175552 }, { "epoch": 0.17, "learning_rate": 0.0004167301905717151, "loss": 3.0444, "theoretical_loss": 3.8554858106431675, "tokens_seen": 577241088 }, { "epoch": 0.17, "learning_rate": 0.00041672016048144436, "loss": 3.0292, "theoretical_loss": 3.8554398280089472, "tokens_seen": 577306624 }, { "epoch": 0.17, "learning_rate": 0.0004167101303911735, "loss": 2.7328, "theoretical_loss": 3.8553938520557915, "tokens_seen": 577372160 }, { "epoch": 0.17, "learning_rate": 0.0004167001003009027, "loss": 2.765, "theoretical_loss": 3.8553478827819707, "tokens_seen": 577437696 }, { "epoch": 0.18, "learning_rate": 0.0004166900702106319, "loss": 2.8839, "theoretical_loss": 3.8553019201857577, "tokens_seen": 577503232 }, { "epoch": 0.18, "learning_rate": 0.0004166800401203611, "loss": 2.7088, "theoretical_loss": 3.8552559642654236, "tokens_seen": 577568768 }, { "epoch": 0.18, "learning_rate": 0.00041667001003009026, "loss": 2.9248, "theoretical_loss": 3.8552100150192423, "tokens_seen": 577634304 }, { "epoch": 0.18, "learning_rate": 0.00041665997993981944, "loss": 2.7546, "theoretical_loss": 3.855164072445487, "tokens_seen": 577699840 }, { "epoch": 0.18, "learning_rate": 0.0004166499498495487, "loss": 3.0814, "theoretical_loss": 3.855118136542433, "tokens_seen": 577765376 }, { "epoch": 0.18, "learning_rate": 0.00041663991975927786, "loss": 2.6923, "theoretical_loss": 3.8550722073083543, "tokens_seen": 577830912 }, { "epoch": 0.18, "learning_rate": 0.00041662988966900704, "loss": 2.7361, "theoretical_loss": 3.855026284741526, "tokens_seen": 577896448 }, { "epoch": 0.18, "learning_rate": 0.0004166198595787362, "loss": 2.867, "theoretical_loss": 3.8549803688402253, "tokens_seen": 577961984 }, { "epoch": 0.18, "learning_rate": 0.0004166098294884654, "loss": 2.765, "theoretical_loss": 3.8549344596027293, "tokens_seen": 578027520 }, { "epoch": 0.18, "learning_rate": 0.0004165997993981946, "loss": 3.1613, "theoretical_loss": 3.854888557027315, "tokens_seen": 578093056 }, { "epoch": 0.18, "learning_rate": 0.0004165897693079238, "loss": 2.8802, "theoretical_loss": 3.8548426611122597, "tokens_seen": 578158592 }, { "epoch": 0.18, "learning_rate": 0.00041657973921765295, "loss": 2.7569, "theoretical_loss": 3.8547967718558436, "tokens_seen": 578224128 }, { "epoch": 0.18, "learning_rate": 0.0004165697091273822, "loss": 2.7319, "theoretical_loss": 3.854750889256345, "tokens_seen": 578289664 }, { "epoch": 0.18, "objective/train/docs_used": 337192, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4397428035736084, "objective/train/theoretical_loss": 3.854705013312045, "objective/train/tokens_used": 598815200, "theoretical_loss": 3.854705013312045, "tokens_seen": 578355200 }, { "epoch": 0.18, "learning_rate": 0.00041655967903711136, "loss": 3.0677, "theoretical_loss": 3.854705013312045, "tokens_seen": 578355200 }, { "epoch": 0.18, "learning_rate": 0.00041654964894684055, "loss": 2.7335, "theoretical_loss": 3.8546591440212237, "tokens_seen": 578420736 }, { "epoch": 0.18, "learning_rate": 0.0004165396188565697, "loss": 2.7197, "theoretical_loss": 3.8546132813821625, "tokens_seen": 578486272 }, { "epoch": 0.18, "learning_rate": 0.0004165295887662989, "loss": 2.7768, "theoretical_loss": 3.8545674253931432, "tokens_seen": 578551808 }, { "epoch": 0.18, "learning_rate": 0.0004165195586760281, "loss": 3.0142, "theoretical_loss": 3.8545215760524494, "tokens_seen": 578617344 }, { "epoch": 0.18, "learning_rate": 0.0004165095285857573, "loss": 2.7591, "theoretical_loss": 3.8544757333583632, "tokens_seen": 578682880 }, { "epoch": 0.18, "learning_rate": 0.00041649949849548645, "loss": 2.8577, "theoretical_loss": 3.854429897309169, "tokens_seen": 578748416 }, { "epoch": 0.18, "learning_rate": 0.0004164894684052157, "loss": 2.8507, "theoretical_loss": 3.854384067903151, "tokens_seen": 578813952 }, { "epoch": 0.18, "learning_rate": 0.0004164794383149448, "loss": 3.0269, "theoretical_loss": 3.854338245138595, "tokens_seen": 578879488 }, { "epoch": 0.18, "learning_rate": 0.00041646940822467405, "loss": 3.0333, "theoretical_loss": 3.8542924290137863, "tokens_seen": 578945024 }, { "epoch": 0.18, "learning_rate": 0.00041645937813440323, "loss": 3.0334, "theoretical_loss": 3.8542466195270118, "tokens_seen": 579010560 }, { "epoch": 0.18, "learning_rate": 0.0004164493480441324, "loss": 2.9967, "theoretical_loss": 3.8542008166765576, "tokens_seen": 579076096 }, { "epoch": 0.18, "learning_rate": 0.0004164393179538616, "loss": 2.7751, "theoretical_loss": 3.8541550204607127, "tokens_seen": 579141632 }, { "epoch": 0.18, "learning_rate": 0.0004164292878635908, "loss": 2.7921, "theoretical_loss": 3.8541092308777647, "tokens_seen": 579207168 }, { "epoch": 0.18, "learning_rate": 0.00041641925777331995, "loss": 2.8874, "theoretical_loss": 3.854063447926003, "tokens_seen": 579272704 }, { "epoch": 0.18, "learning_rate": 0.0004164092276830492, "loss": 3.0442, "theoretical_loss": 3.8540176716037164, "tokens_seen": 579338240 }, { "epoch": 0.18, "learning_rate": 0.0004163991975927783, "loss": 2.8858, "theoretical_loss": 3.853971901909196, "tokens_seen": 579403776 }, { "epoch": 0.18, "learning_rate": 0.00041638916750250755, "loss": 2.8424, "theoretical_loss": 3.8539261388407327, "tokens_seen": 579469312 }, { "epoch": 0.18, "learning_rate": 0.00041637913741223673, "loss": 2.9964, "theoretical_loss": 3.8538803823966172, "tokens_seen": 579534848 }, { "epoch": 0.18, "learning_rate": 0.0004163691073219659, "loss": 2.843, "theoretical_loss": 3.853834632575143, "tokens_seen": 579600384 }, { "epoch": 0.18, "learning_rate": 0.0004163590772316951, "loss": 2.793, "theoretical_loss": 3.8537888893746013, "tokens_seen": 579665920 }, { "epoch": 0.18, "learning_rate": 0.0004163490471414243, "loss": 3.0142, "theoretical_loss": 3.853743152793287, "tokens_seen": 579731456 }, { "epoch": 0.18, "learning_rate": 0.00041633901705115346, "loss": 2.8166, "theoretical_loss": 3.8536974228294927, "tokens_seen": 579796992 }, { "epoch": 0.18, "learning_rate": 0.0004163289869608827, "loss": 2.996, "theoretical_loss": 3.8536516994815146, "tokens_seen": 579862528 }, { "epoch": 0.18, "learning_rate": 0.0004163189568706118, "loss": 2.8994, "theoretical_loss": 3.8536059827476477, "tokens_seen": 579928064 }, { "epoch": 0.18, "objective/train/docs_used": 337878, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1601502895355225, "objective/train/theoretical_loss": 3.8535602726261864, "objective/train/tokens_used": 600453600, "theoretical_loss": 3.8535602726261864, "tokens_seen": 579993600 }, { "epoch": 0.18, "learning_rate": 0.00041630892678034105, "loss": 2.8386, "theoretical_loss": 3.8535602726261864, "tokens_seen": 579993600 }, { "epoch": 0.18, "learning_rate": 0.0004162988966900702, "loss": 2.8647, "theoretical_loss": 3.853514569115429, "tokens_seen": 580059136 }, { "epoch": 0.18, "learning_rate": 0.0004162888665997994, "loss": 2.9424, "theoretical_loss": 3.8534688722136723, "tokens_seen": 580124672 }, { "epoch": 0.18, "learning_rate": 0.0004162788365095286, "loss": 2.7317, "theoretical_loss": 3.853423181919214, "tokens_seen": 580190208 }, { "epoch": 0.18, "learning_rate": 0.0004162688064192578, "loss": 2.8081, "theoretical_loss": 3.8533774982303526, "tokens_seen": 580255744 }, { "epoch": 0.18, "learning_rate": 0.00041625877632898696, "loss": 3.023, "theoretical_loss": 3.853331821145387, "tokens_seen": 580321280 }, { "epoch": 0.18, "learning_rate": 0.0004162487462387162, "loss": 2.8905, "theoretical_loss": 3.8532861506626173, "tokens_seen": 580386816 }, { "epoch": 0.18, "learning_rate": 0.0004162387161484453, "loss": 2.8099, "theoretical_loss": 3.853240486780344, "tokens_seen": 580452352 }, { "epoch": 0.18, "learning_rate": 0.00041622868605817456, "loss": 3.0392, "theoretical_loss": 3.8531948294968674, "tokens_seen": 580517888 }, { "epoch": 0.18, "learning_rate": 0.0004162186559679037, "loss": 2.6887, "theoretical_loss": 3.85314917881049, "tokens_seen": 580583424 }, { "epoch": 0.18, "learning_rate": 0.0004162086258776329, "loss": 2.8255, "theoretical_loss": 3.853103534719513, "tokens_seen": 580648960 }, { "epoch": 0.18, "learning_rate": 0.0004161985957873621, "loss": 2.6804, "theoretical_loss": 3.8530578972222402, "tokens_seen": 580714496 }, { "epoch": 0.18, "learning_rate": 0.0004161885656970913, "loss": 3.0892, "theoretical_loss": 3.8530122663169752, "tokens_seen": 580780032 }, { "epoch": 0.18, "learning_rate": 0.00041617853560682046, "loss": 2.7583, "theoretical_loss": 3.8529666420020217, "tokens_seen": 580845568 }, { "epoch": 0.18, "learning_rate": 0.00041616850551654964, "loss": 2.6621, "theoretical_loss": 3.852921024275684, "tokens_seen": 580911104 }, { "epoch": 0.18, "learning_rate": 0.0004161584754262788, "loss": 2.7139, "theoretical_loss": 3.8528754131362684, "tokens_seen": 580976640 }, { "epoch": 0.18, "learning_rate": 0.00041614844533600806, "loss": 3.1473, "theoretical_loss": 3.8528298085820802, "tokens_seen": 581042176 }, { "epoch": 0.18, "learning_rate": 0.0004161384152457372, "loss": 3.281, "theoretical_loss": 3.8527842106114267, "tokens_seen": 581107712 }, { "epoch": 0.18, "learning_rate": 0.0004161283851554664, "loss": 2.8712, "theoretical_loss": 3.852738619222615, "tokens_seen": 581173248 }, { "epoch": 0.18, "learning_rate": 0.00041611835506519555, "loss": 2.9122, "theoretical_loss": 3.8526930344139525, "tokens_seen": 581238784 }, { "epoch": 0.18, "learning_rate": 0.0004161083249749248, "loss": 2.7869, "theoretical_loss": 3.8526474561837483, "tokens_seen": 581304320 }, { "epoch": 0.18, "learning_rate": 0.00041609829488465397, "loss": 2.9412, "theoretical_loss": 3.852601884530311, "tokens_seen": 581369856 }, { "epoch": 0.18, "learning_rate": 0.00041608826479438315, "loss": 2.6746, "theoretical_loss": 3.852556319451952, "tokens_seen": 581435392 }, { "epoch": 0.18, "learning_rate": 0.00041607823470411233, "loss": 3.056, "theoretical_loss": 3.8525107609469793, "tokens_seen": 581500928 }, { "epoch": 0.18, "learning_rate": 0.00041606820461384156, "loss": 2.8142, "theoretical_loss": 3.852465209013705, "tokens_seen": 581566464 }, { "epoch": 0.18, "objective/train/docs_used": 339259, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0680527687072754, "objective/train/theoretical_loss": 3.8524196636504415, "objective/train/tokens_used": 602092000, "theoretical_loss": 3.8524196636504415, "tokens_seen": 581632000 }, { "epoch": 0.18, "learning_rate": 0.0004160581745235707, "loss": 2.9815, "theoretical_loss": 3.8524196636504415, "tokens_seen": 581632000 }, { "epoch": 0.18, "learning_rate": 0.0004160481444332999, "loss": 2.8625, "theoretical_loss": 3.8523741248555, "tokens_seen": 581697536 }, { "epoch": 0.18, "learning_rate": 0.00041603811434302905, "loss": 2.9147, "theoretical_loss": 3.8523285926271935, "tokens_seen": 581763072 }, { "epoch": 0.18, "learning_rate": 0.0004160280842527583, "loss": 2.8707, "theoretical_loss": 3.852283066963836, "tokens_seen": 581828608 }, { "epoch": 0.18, "learning_rate": 0.00041601805416248747, "loss": 3.0122, "theoretical_loss": 3.8522375478637425, "tokens_seen": 581894144 }, { "epoch": 0.18, "learning_rate": 0.00041600802407221665, "loss": 2.7454, "theoretical_loss": 3.852192035325226, "tokens_seen": 581959680 }, { "epoch": 0.18, "learning_rate": 0.00041599799398194583, "loss": 2.6902, "theoretical_loss": 3.8521465293466024, "tokens_seen": 582025216 }, { "epoch": 0.18, "learning_rate": 0.000415987963891675, "loss": 2.7896, "theoretical_loss": 3.852101029926188, "tokens_seen": 582090752 }, { "epoch": 0.18, "learning_rate": 0.0004159779338014042, "loss": 2.8443, "theoretical_loss": 3.8520555370622995, "tokens_seen": 582156288 }, { "epoch": 0.18, "learning_rate": 0.00041596790371113343, "loss": 2.575, "theoretical_loss": 3.852010050753254, "tokens_seen": 582221824 }, { "epoch": 0.18, "learning_rate": 0.00041595787362086256, "loss": 2.7256, "theoretical_loss": 3.85196457099737, "tokens_seen": 582287360 }, { "epoch": 0.18, "learning_rate": 0.0004159478435305918, "loss": 2.8148, "theoretical_loss": 3.8519190977929645, "tokens_seen": 582352896 }, { "epoch": 0.18, "learning_rate": 0.0004159378134403209, "loss": 2.7325, "theoretical_loss": 3.851873631138358, "tokens_seen": 582418432 }, { "epoch": 0.18, "learning_rate": 0.00041592778335005015, "loss": 2.8538, "theoretical_loss": 3.85182817103187, "tokens_seen": 582483968 }, { "epoch": 0.18, "learning_rate": 0.00041591775325977934, "loss": 2.8132, "theoretical_loss": 3.8517827174718198, "tokens_seen": 582549504 }, { "epoch": 0.18, "learning_rate": 0.0004159077231695085, "loss": 2.6763, "theoretical_loss": 3.85173727045653, "tokens_seen": 582615040 }, { "epoch": 0.18, "learning_rate": 0.00041589769307923775, "loss": 2.7962, "theoretical_loss": 3.851691829984321, "tokens_seen": 582680576 }, { "epoch": 0.18, "learning_rate": 0.00041588766298896693, "loss": 2.82, "theoretical_loss": 3.851646396053516, "tokens_seen": 582746112 }, { "epoch": 0.18, "learning_rate": 0.0004158776328986961, "loss": 2.9067, "theoretical_loss": 3.851600968662437, "tokens_seen": 582811648 }, { "epoch": 0.18, "learning_rate": 0.0004158676028084253, "loss": 2.6858, "theoretical_loss": 3.851555547809408, "tokens_seen": 582877184 }, { "epoch": 0.18, "learning_rate": 0.0004158575727181545, "loss": 2.7076, "theoretical_loss": 3.851510133492752, "tokens_seen": 582942720 }, { "epoch": 0.18, "learning_rate": 0.00041584754262788366, "loss": 2.7853, "theoretical_loss": 3.8514647257107955, "tokens_seen": 583008256 }, { "epoch": 0.18, "learning_rate": 0.0004158375125376129, "loss": 3.0245, "theoretical_loss": 3.8514193244618617, "tokens_seen": 583073792 }, { "epoch": 0.18, "learning_rate": 0.000415827482447342, "loss": 2.8126, "theoretical_loss": 3.8513739297442786, "tokens_seen": 583139328 }, { "epoch": 0.18, "learning_rate": 0.00041581745235707126, "loss": 2.9131, "theoretical_loss": 3.851328541556372, "tokens_seen": 583204864 }, { "epoch": 0.18, "objective/train/docs_used": 340083, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4392688274383545, "objective/train/theoretical_loss": 3.851283159896468, "objective/train/tokens_used": 603730400, "theoretical_loss": 3.851283159896468, "tokens_seen": 583270400 }, { "epoch": 0.18, "learning_rate": 0.0004158074222668004, "loss": 2.9399, "theoretical_loss": 3.851283159896468, "tokens_seen": 583270400 }, { "epoch": 0.18, "learning_rate": 0.0004157973921765296, "loss": 2.9062, "theoretical_loss": 3.851237784762896, "tokens_seen": 583335936 }, { "epoch": 0.18, "learning_rate": 0.0004157873620862588, "loss": 2.7214, "theoretical_loss": 3.8511924161539826, "tokens_seen": 583401472 }, { "epoch": 0.18, "learning_rate": 0.000415777331995988, "loss": 2.7425, "theoretical_loss": 3.8511470540680586, "tokens_seen": 583467008 }, { "epoch": 0.18, "learning_rate": 0.00041576730190571716, "loss": 2.8637, "theoretical_loss": 3.8511016985034523, "tokens_seen": 583532544 }, { "epoch": 0.18, "learning_rate": 0.0004157572718154464, "loss": 2.9176, "theoretical_loss": 3.851056349458495, "tokens_seen": 583598080 }, { "epoch": 0.18, "learning_rate": 0.0004157472417251755, "loss": 2.9171, "theoretical_loss": 3.8510110069315173, "tokens_seen": 583663616 }, { "epoch": 0.18, "learning_rate": 0.00041573721163490476, "loss": 2.9357, "theoretical_loss": 3.85096567092085, "tokens_seen": 583729152 }, { "epoch": 0.18, "learning_rate": 0.0004157271815446339, "loss": 2.7137, "theoretical_loss": 3.8509203414248256, "tokens_seen": 583794688 }, { "epoch": 0.18, "learning_rate": 0.0004157171514543631, "loss": 2.7805, "theoretical_loss": 3.8508750184417764, "tokens_seen": 583860224 }, { "epoch": 0.18, "learning_rate": 0.0004157071213640923, "loss": 2.8358, "theoretical_loss": 3.850829701970037, "tokens_seen": 583925760 }, { "epoch": 0.18, "learning_rate": 0.0004156970912738215, "loss": 2.943, "theoretical_loss": 3.85078439200794, "tokens_seen": 583991296 }, { "epoch": 0.18, "learning_rate": 0.00041568706118355066, "loss": 2.6781, "theoretical_loss": 3.85073908855382, "tokens_seen": 584056832 }, { "epoch": 0.18, "learning_rate": 0.00041567703109327984, "loss": 2.738, "theoretical_loss": 3.8506937916060133, "tokens_seen": 584122368 }, { "epoch": 0.18, "learning_rate": 0.000415667001003009, "loss": 2.8833, "theoretical_loss": 3.8506485011628544, "tokens_seen": 584187904 }, { "epoch": 0.18, "learning_rate": 0.00041565697091273826, "loss": 3.0191, "theoretical_loss": 3.8506032172226803, "tokens_seen": 584253440 }, { "epoch": 0.18, "learning_rate": 0.0004156469408224674, "loss": 2.5674, "theoretical_loss": 3.850557939783828, "tokens_seen": 584318976 }, { "epoch": 0.18, "learning_rate": 0.0004156369107321966, "loss": 2.6857, "theoretical_loss": 3.850512668844635, "tokens_seen": 584384512 }, { "epoch": 0.18, "learning_rate": 0.00041562688064192575, "loss": 2.6629, "theoretical_loss": 3.8504674044034393, "tokens_seen": 584450048 }, { "epoch": 0.18, "learning_rate": 0.000415616850551655, "loss": 2.6449, "theoretical_loss": 3.8504221464585795, "tokens_seen": 584515584 }, { "epoch": 0.18, "learning_rate": 0.00041560682046138417, "loss": 2.8412, "theoretical_loss": 3.850376895008396, "tokens_seen": 584581120 }, { "epoch": 0.18, "learning_rate": 0.00041559679037111335, "loss": 3.0066, "theoretical_loss": 3.8503316500512286, "tokens_seen": 584646656 }, { "epoch": 0.18, "learning_rate": 0.00041558676028084253, "loss": 2.8835, "theoretical_loss": 3.850286411585417, "tokens_seen": 584712192 }, { "epoch": 0.18, "learning_rate": 0.00041557673019057176, "loss": 2.687, "theoretical_loss": 3.8502411796093035, "tokens_seen": 584777728 }, { "epoch": 0.18, "learning_rate": 0.0004155667001003009, "loss": 2.865, "theoretical_loss": 3.8501959541212294, "tokens_seen": 584843264 }, { "epoch": 0.18, "objective/train/docs_used": 341398, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5760111808776855, "objective/train/theoretical_loss": 3.850150735119537, "objective/train/tokens_used": 605368800, "theoretical_loss": 3.850150735119537, "tokens_seen": 584908800 }, { "epoch": 0.18, "learning_rate": 0.0004155566700100301, "loss": 2.7967, "theoretical_loss": 3.850150735119537, "tokens_seen": 584908800 }, { "epoch": 0.18, "learning_rate": 0.00041554663991975925, "loss": 2.7677, "theoretical_loss": 3.8501055226025707, "tokens_seen": 584974336 }, { "epoch": 0.18, "learning_rate": 0.0004155366098294885, "loss": 2.9202, "theoretical_loss": 3.8500603165686726, "tokens_seen": 585039872 }, { "epoch": 0.18, "learning_rate": 0.00041552657973921767, "loss": 2.9313, "theoretical_loss": 3.850015117016188, "tokens_seen": 585105408 }, { "epoch": 0.18, "learning_rate": 0.00041551654964894685, "loss": 2.7369, "theoretical_loss": 3.849969923943461, "tokens_seen": 585170944 }, { "epoch": 0.18, "learning_rate": 0.00041550651955867603, "loss": 2.5288, "theoretical_loss": 3.8499247373488377, "tokens_seen": 585236480 }, { "epoch": 0.18, "learning_rate": 0.0004154964894684052, "loss": 2.936, "theoretical_loss": 3.8498795572306643, "tokens_seen": 585302016 }, { "epoch": 0.18, "learning_rate": 0.0004154864593781344, "loss": 2.9509, "theoretical_loss": 3.8498343835872872, "tokens_seen": 585367552 }, { "epoch": 0.18, "learning_rate": 0.00041547642928786363, "loss": 2.8473, "theoretical_loss": 3.8497892164170544, "tokens_seen": 585433088 }, { "epoch": 0.18, "learning_rate": 0.00041546639919759276, "loss": 2.6784, "theoretical_loss": 3.8497440557183134, "tokens_seen": 585498624 }, { "epoch": 0.18, "learning_rate": 0.000415456369107322, "loss": 2.5299, "theoretical_loss": 3.849698901489412, "tokens_seen": 585564160 }, { "epoch": 0.18, "learning_rate": 0.0004154463390170511, "loss": 2.8405, "theoretical_loss": 3.849653753728701, "tokens_seen": 585629696 }, { "epoch": 0.18, "learning_rate": 0.00041543630892678035, "loss": 2.9255, "theoretical_loss": 3.849608612434529, "tokens_seen": 585695232 }, { "epoch": 0.18, "learning_rate": 0.00041542627883650954, "loss": 2.8259, "theoretical_loss": 3.8495634776052463, "tokens_seen": 585760768 }, { "epoch": 0.18, "learning_rate": 0.0004154162487462387, "loss": 2.873, "theoretical_loss": 3.849518349239205, "tokens_seen": 585826304 }, { "epoch": 0.18, "learning_rate": 0.0004154062186559679, "loss": 3.1558, "theoretical_loss": 3.849473227334755, "tokens_seen": 585891840 }, { "epoch": 0.18, "learning_rate": 0.00041539618856569713, "loss": 2.99, "theoretical_loss": 3.8494281118902505, "tokens_seen": 585957376 }, { "epoch": 0.18, "learning_rate": 0.00041538615847542626, "loss": 2.9531, "theoretical_loss": 3.849383002904043, "tokens_seen": 586022912 }, { "epoch": 0.18, "learning_rate": 0.0004153761283851555, "loss": 2.7522, "theoretical_loss": 3.8493379003744854, "tokens_seen": 586088448 }, { "epoch": 0.18, "learning_rate": 0.0004153660982948846, "loss": 2.6776, "theoretical_loss": 3.8492928042999335, "tokens_seen": 586153984 }, { "epoch": 0.18, "learning_rate": 0.00041535606820461386, "loss": 2.7751, "theoretical_loss": 3.8492477146787403, "tokens_seen": 586219520 }, { "epoch": 0.18, "learning_rate": 0.00041534603811434304, "loss": 2.825, "theoretical_loss": 3.849202631509262, "tokens_seen": 586285056 }, { "epoch": 0.18, "learning_rate": 0.0004153360080240722, "loss": 2.8614, "theoretical_loss": 3.849157554789853, "tokens_seen": 586350592 }, { "epoch": 0.18, "learning_rate": 0.0004153259779338014, "loss": 2.9808, "theoretical_loss": 3.8491124845188716, "tokens_seen": 586416128 }, { "epoch": 0.18, "learning_rate": 0.0004153159478435306, "loss": 2.8579, "theoretical_loss": 3.8490674206946736, "tokens_seen": 586481664 }, { "epoch": 0.18, "objective/train/docs_used": 342071, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.825225353240967, "objective/train/theoretical_loss": 3.8490223633156173, "objective/train/tokens_used": 607007200, "theoretical_loss": 3.8490223633156173, "tokens_seen": 586547200 }, { "epoch": 0.18, "learning_rate": 0.00041530591775325976, "loss": 2.7571, "theoretical_loss": 3.8490223633156173, "tokens_seen": 586547200 }, { "epoch": 0.18, "learning_rate": 0.000415295887662989, "loss": 2.7614, "theoretical_loss": 3.84897731238006, "tokens_seen": 586612736 }, { "epoch": 0.18, "learning_rate": 0.0004152858575727181, "loss": 2.8002, "theoretical_loss": 3.8489322678863624, "tokens_seen": 586678272 }, { "epoch": 0.18, "learning_rate": 0.00041527582748244736, "loss": 2.8835, "theoretical_loss": 3.8488872298328816, "tokens_seen": 586743808 }, { "epoch": 0.18, "learning_rate": 0.0004152657973921765, "loss": 2.8548, "theoretical_loss": 3.848842198217979, "tokens_seen": 586809344 }, { "epoch": 0.18, "learning_rate": 0.0004152557673019057, "loss": 2.9912, "theoretical_loss": 3.8487971730400146, "tokens_seen": 586874880 }, { "epoch": 0.18, "learning_rate": 0.0004152457372116349, "loss": 2.8354, "theoretical_loss": 3.84875215429735, "tokens_seen": 586940416 }, { "epoch": 0.18, "learning_rate": 0.0004152357071213641, "loss": 2.9608, "theoretical_loss": 3.8487071419883474, "tokens_seen": 587005952 }, { "epoch": 0.18, "learning_rate": 0.00041522567703109327, "loss": 2.9313, "theoretical_loss": 3.8486621361113684, "tokens_seen": 587071488 }, { "epoch": 0.18, "learning_rate": 0.0004152156469408225, "loss": 2.9621, "theoretical_loss": 3.848617136664777, "tokens_seen": 587137024 }, { "epoch": 0.18, "learning_rate": 0.00041520561685055163, "loss": 2.762, "theoretical_loss": 3.848572143646936, "tokens_seen": 587202560 }, { "epoch": 0.18, "learning_rate": 0.00041519558676028086, "loss": 2.8418, "theoretical_loss": 3.8485271570562096, "tokens_seen": 587268096 }, { "epoch": 0.18, "learning_rate": 0.00041518555667001, "loss": 2.9505, "theoretical_loss": 3.8484821768909634, "tokens_seen": 587333632 }, { "epoch": 0.18, "learning_rate": 0.0004151755265797392, "loss": 2.7378, "theoretical_loss": 3.848437203149562, "tokens_seen": 587399168 }, { "epoch": 0.18, "learning_rate": 0.0004151654964894684, "loss": 2.7264, "theoretical_loss": 3.8483922358303717, "tokens_seen": 587464704 }, { "epoch": 0.18, "learning_rate": 0.0004151554663991976, "loss": 2.6414, "theoretical_loss": 3.8483472749317595, "tokens_seen": 587530240 }, { "epoch": 0.18, "learning_rate": 0.0004151454363089268, "loss": 2.7191, "theoretical_loss": 3.8483023204520928, "tokens_seen": 587595776 }, { "epoch": 0.18, "learning_rate": 0.00041513540621865595, "loss": 2.5959, "theoretical_loss": 3.848257372389738, "tokens_seen": 587661312 }, { "epoch": 0.18, "learning_rate": 0.0004151253761283852, "loss": 2.9658, "theoretical_loss": 3.848212430743065, "tokens_seen": 587726848 }, { "epoch": 0.18, "learning_rate": 0.00041511534603811437, "loss": 2.544, "theoretical_loss": 3.8481674955104426, "tokens_seen": 587792384 }, { "epoch": 0.18, "learning_rate": 0.00041510531594784355, "loss": 2.8289, "theoretical_loss": 3.848122566690239, "tokens_seen": 587857920 }, { "epoch": 0.18, "learning_rate": 0.00041509528585757273, "loss": 2.7469, "theoretical_loss": 3.8480776442808264, "tokens_seen": 587923456 }, { "epoch": 0.18, "learning_rate": 0.00041508525576730196, "loss": 2.9854, "theoretical_loss": 3.8480327282805744, "tokens_seen": 587988992 }, { "epoch": 0.18, "learning_rate": 0.0004150752256770311, "loss": 3.1744, "theoretical_loss": 3.8479878186878547, "tokens_seen": 588054528 }, { "epoch": 0.18, "learning_rate": 0.0004150651955867603, "loss": 3.0129, "theoretical_loss": 3.847942915501039, "tokens_seen": 588120064 }, { "epoch": 0.18, "objective/train/docs_used": 342650, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.821354866027832, "objective/train/theoretical_loss": 3.8478980187185003, "objective/train/tokens_used": 608645600, "theoretical_loss": 3.8478980187185003, "tokens_seen": 588185600 }, { "epoch": 0.18, "learning_rate": 0.00041505516549648945, "loss": 2.734, "theoretical_loss": 3.8478980187185003, "tokens_seen": 588185600 }, { "epoch": 0.18, "learning_rate": 0.0004150451354062187, "loss": 2.91, "theoretical_loss": 3.8478531283386124, "tokens_seen": 588251136 }, { "epoch": 0.18, "learning_rate": 0.00041503510531594787, "loss": 2.9527, "theoretical_loss": 3.847808244359747, "tokens_seen": 588316672 }, { "epoch": 0.18, "learning_rate": 0.00041502507522567705, "loss": 2.9137, "theoretical_loss": 3.847763366780281, "tokens_seen": 588382208 }, { "epoch": 0.18, "learning_rate": 0.00041501504513540623, "loss": 2.8705, "theoretical_loss": 3.8477184955985875, "tokens_seen": 588447744 }, { "epoch": 0.18, "learning_rate": 0.0004150050150451354, "loss": 2.7977, "theoretical_loss": 3.8476736308130426, "tokens_seen": 588513280 }, { "epoch": 0.18, "learning_rate": 0.0004149949849548646, "loss": 2.9383, "theoretical_loss": 3.8476287724220226, "tokens_seen": 588578816 }, { "epoch": 0.18, "learning_rate": 0.00041498495486459383, "loss": 2.7924, "theoretical_loss": 3.8475839204239044, "tokens_seen": 588644352 }, { "epoch": 0.18, "learning_rate": 0.00041497492477432296, "loss": 2.854, "theoretical_loss": 3.847539074817065, "tokens_seen": 588709888 }, { "epoch": 0.18, "learning_rate": 0.0004149648946840522, "loss": 2.9575, "theoretical_loss": 3.8474942355998825, "tokens_seen": 588775424 }, { "epoch": 0.18, "learning_rate": 0.0004149548645937813, "loss": 2.7192, "theoretical_loss": 3.847449402770735, "tokens_seen": 588840960 }, { "epoch": 0.18, "learning_rate": 0.00041494483450351055, "loss": 2.7516, "theoretical_loss": 3.8474045763280023, "tokens_seen": 588906496 }, { "epoch": 0.18, "learning_rate": 0.00041493480441323974, "loss": 2.9051, "theoretical_loss": 3.847359756270064, "tokens_seen": 588972032 }, { "epoch": 0.18, "learning_rate": 0.0004149247743229689, "loss": 2.822, "theoretical_loss": 3.8473149425953, "tokens_seen": 589037568 }, { "epoch": 0.18, "learning_rate": 0.0004149147442326981, "loss": 2.9753, "theoretical_loss": 3.8472701353020913, "tokens_seen": 589103104 }, { "epoch": 0.18, "learning_rate": 0.00041490471414242733, "loss": 2.5136, "theoretical_loss": 3.8472253343888196, "tokens_seen": 589168640 }, { "epoch": 0.18, "learning_rate": 0.00041489468405215646, "loss": 2.9819, "theoretical_loss": 3.8471805398538663, "tokens_seen": 589234176 }, { "epoch": 0.18, "learning_rate": 0.0004148846539618857, "loss": 3.0167, "theoretical_loss": 3.8471357516956153, "tokens_seen": 589299712 }, { "epoch": 0.18, "learning_rate": 0.0004148746238716148, "loss": 3.0563, "theoretical_loss": 3.8470909699124487, "tokens_seen": 589365248 }, { "epoch": 0.18, "learning_rate": 0.00041486459378134406, "loss": 2.8984, "theoretical_loss": 3.8470461945027505, "tokens_seen": 589430784 }, { "epoch": 0.18, "learning_rate": 0.00041485456369107324, "loss": 2.9522, "theoretical_loss": 3.8470014254649056, "tokens_seen": 589496320 }, { "epoch": 0.18, "learning_rate": 0.0004148445336008024, "loss": 2.7076, "theoretical_loss": 3.8469566627972993, "tokens_seen": 589561856 }, { "epoch": 0.18, "learning_rate": 0.0004148345035105316, "loss": 2.6847, "theoretical_loss": 3.846911906498316, "tokens_seen": 589627392 }, { "epoch": 0.18, "learning_rate": 0.0004148244734202608, "loss": 2.6315, "theoretical_loss": 3.8468671565663426, "tokens_seen": 589692928 }, { "epoch": 0.18, "learning_rate": 0.00041481444332998996, "loss": 2.9967, "theoretical_loss": 3.8468224129997663, "tokens_seen": 589758464 }, { "epoch": 0.18, "objective/train/docs_used": 343935, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.133723497390747, "objective/train/theoretical_loss": 3.846777675796974, "objective/train/tokens_used": 610284000, "theoretical_loss": 3.846777675796974, "tokens_seen": 589824000 }, { "epoch": 0.18, "learning_rate": 0.0004148044132397192, "loss": 2.9625, "theoretical_loss": 3.846777675796974, "tokens_seen": 589824000 }, { "epoch": 0.18, "learning_rate": 0.0004147943831494483, "loss": 2.7877, "theoretical_loss": 3.8467329449563534, "tokens_seen": 589889536 }, { "epoch": 0.18, "learning_rate": 0.00041478435305917756, "loss": 2.6622, "theoretical_loss": 3.8466882204762936, "tokens_seen": 589955072 }, { "epoch": 0.18, "learning_rate": 0.0004147743229689067, "loss": 2.7714, "theoretical_loss": 3.846643502355183, "tokens_seen": 590020608 }, { "epoch": 0.18, "learning_rate": 0.0004147642928786359, "loss": 2.9137, "theoretical_loss": 3.846598790591412, "tokens_seen": 590086144 }, { "epoch": 0.18, "learning_rate": 0.0004147542627883651, "loss": 3.0009, "theoretical_loss": 3.846554085183371, "tokens_seen": 590151680 }, { "epoch": 0.18, "learning_rate": 0.0004147442326980943, "loss": 2.5204, "theoretical_loss": 3.8465093861294504, "tokens_seen": 590217216 }, { "epoch": 0.18, "learning_rate": 0.00041473420260782347, "loss": 2.6391, "theoretical_loss": 3.8464646934280413, "tokens_seen": 590282752 }, { "epoch": 0.18, "learning_rate": 0.0004147241725175527, "loss": 2.7687, "theoretical_loss": 3.846420007077537, "tokens_seen": 590348288 }, { "epoch": 0.18, "learning_rate": 0.00041471414242728183, "loss": 2.841, "theoretical_loss": 3.8463753270763297, "tokens_seen": 590413824 }, { "epoch": 0.18, "learning_rate": 0.00041470411233701106, "loss": 2.8472, "theoretical_loss": 3.846330653422812, "tokens_seen": 590479360 }, { "epoch": 0.18, "learning_rate": 0.0004146940822467402, "loss": 2.7101, "theoretical_loss": 3.8462859861153778, "tokens_seen": 590544896 }, { "epoch": 0.18, "learning_rate": 0.0004146840521564694, "loss": 2.6634, "theoretical_loss": 3.8462413251524215, "tokens_seen": 590610432 }, { "epoch": 0.18, "learning_rate": 0.0004146740220661986, "loss": 3.0439, "theoretical_loss": 3.846196670532339, "tokens_seen": 590675968 }, { "epoch": 0.18, "learning_rate": 0.0004146639919759278, "loss": 3.0207, "theoretical_loss": 3.8461520222535253, "tokens_seen": 590741504 }, { "epoch": 0.18, "learning_rate": 0.00041465396188565697, "loss": 2.7321, "theoretical_loss": 3.8461073803143764, "tokens_seen": 590807040 }, { "epoch": 0.18, "learning_rate": 0.00041464393179538615, "loss": 2.7399, "theoretical_loss": 3.846062744713289, "tokens_seen": 590872576 }, { "epoch": 0.18, "learning_rate": 0.00041463390170511533, "loss": 2.7948, "theoretical_loss": 3.8460181154486612, "tokens_seen": 590938112 }, { "epoch": 0.18, "learning_rate": 0.00041462387161484457, "loss": 2.9129, "theoretical_loss": 3.8459734925188895, "tokens_seen": 591003648 }, { "epoch": 0.18, "learning_rate": 0.0004146138415245737, "loss": 2.8917, "theoretical_loss": 3.8459288759223735, "tokens_seen": 591069184 }, { "epoch": 0.18, "learning_rate": 0.00041460381143430293, "loss": 2.8784, "theoretical_loss": 3.8458842656575123, "tokens_seen": 591134720 }, { "epoch": 0.18, "learning_rate": 0.0004145937813440321, "loss": 2.9389, "theoretical_loss": 3.845839661722705, "tokens_seen": 591200256 }, { "epoch": 0.18, "learning_rate": 0.0004145837512537613, "loss": 3.1169, "theoretical_loss": 3.8457950641163516, "tokens_seen": 591265792 }, { "epoch": 0.18, "learning_rate": 0.0004145737211634905, "loss": 2.958, "theoretical_loss": 3.845750472836854, "tokens_seen": 591331328 }, { "epoch": 0.18, "learning_rate": 0.00041456369107321965, "loss": 2.7063, "theoretical_loss": 3.845705887882612, "tokens_seen": 591396864 }, { "epoch": 0.18, "objective/train/docs_used": 344744, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.702831983566284, "objective/train/theoretical_loss": 3.845661309252029, "objective/train/tokens_used": 611922400, "theoretical_loss": 3.845661309252029, "tokens_seen": 591462400 }, { "epoch": 0.18, "learning_rate": 0.00041455366098294884, "loss": 2.7622, "theoretical_loss": 3.845661309252029, "tokens_seen": 591462400 }, { "epoch": 0.18, "learning_rate": 0.00041454363089267807, "loss": 2.7417, "theoretical_loss": 3.8456167369435073, "tokens_seen": 591527936 }, { "epoch": 0.18, "learning_rate": 0.0004145336008024072, "loss": 2.776, "theoretical_loss": 3.8455721709554496, "tokens_seen": 591593472 }, { "epoch": 0.18, "learning_rate": 0.00041452357071213643, "loss": 2.8246, "theoretical_loss": 3.84552761128626, "tokens_seen": 591659008 }, { "epoch": 0.18, "learning_rate": 0.00041451354062186556, "loss": 3.0347, "theoretical_loss": 3.8454830579343424, "tokens_seen": 591724544 }, { "epoch": 0.18, "learning_rate": 0.0004145035105315948, "loss": 2.9839, "theoretical_loss": 3.845438510898102, "tokens_seen": 591790080 }, { "epoch": 0.18, "learning_rate": 0.000414493480441324, "loss": 2.7507, "theoretical_loss": 3.8453939701759436, "tokens_seen": 591855616 }, { "epoch": 0.18, "learning_rate": 0.00041448345035105316, "loss": 2.8606, "theoretical_loss": 3.8453494357662747, "tokens_seen": 591921152 }, { "epoch": 0.18, "learning_rate": 0.00041447342026078234, "loss": 3.0087, "theoretical_loss": 3.8453049076675008, "tokens_seen": 591986688 }, { "epoch": 0.18, "learning_rate": 0.0004144633901705115, "loss": 2.8523, "theoretical_loss": 3.8452603858780288, "tokens_seen": 592052224 }, { "epoch": 0.18, "learning_rate": 0.0004144533600802407, "loss": 2.7618, "theoretical_loss": 3.8452158703962676, "tokens_seen": 592117760 }, { "epoch": 0.18, "learning_rate": 0.00041444332998996994, "loss": 2.8818, "theoretical_loss": 3.8451713612206246, "tokens_seen": 592183296 }, { "epoch": 0.18, "learning_rate": 0.00041443329989969906, "loss": 2.8679, "theoretical_loss": 3.8451268583495084, "tokens_seen": 592248832 }, { "epoch": 0.18, "learning_rate": 0.0004144232698094283, "loss": 3.0313, "theoretical_loss": 3.84508236178133, "tokens_seen": 592314368 }, { "epoch": 0.18, "learning_rate": 0.00041441323971915753, "loss": 2.8172, "theoretical_loss": 3.8450378715144984, "tokens_seen": 592379904 }, { "epoch": 0.18, "learning_rate": 0.00041440320962888666, "loss": 2.6686, "theoretical_loss": 3.844993387547425, "tokens_seen": 592445440 }, { "epoch": 0.18, "learning_rate": 0.0004143931795386159, "loss": 2.719, "theoretical_loss": 3.8449489098785197, "tokens_seen": 592510976 }, { "epoch": 0.18, "learning_rate": 0.000414383149448345, "loss": 2.9581, "theoretical_loss": 3.8449044385061955, "tokens_seen": 592576512 }, { "epoch": 0.18, "learning_rate": 0.00041437311935807426, "loss": 2.6012, "theoretical_loss": 3.8448599734288647, "tokens_seen": 592642048 }, { "epoch": 0.18, "learning_rate": 0.00041436308926780344, "loss": 2.8978, "theoretical_loss": 3.8448155146449396, "tokens_seen": 592707584 }, { "epoch": 0.18, "learning_rate": 0.0004143530591775326, "loss": 3.0502, "theoretical_loss": 3.844771062152834, "tokens_seen": 592773120 }, { "epoch": 0.18, "learning_rate": 0.0004143430290872618, "loss": 2.8521, "theoretical_loss": 3.8447266159509628, "tokens_seen": 592838656 }, { "epoch": 0.18, "learning_rate": 0.000414332998996991, "loss": 2.7796, "theoretical_loss": 3.8446821760377397, "tokens_seen": 592904192 }, { "epoch": 0.18, "learning_rate": 0.00041432296890672016, "loss": 2.7843, "theoretical_loss": 3.84463774241158, "tokens_seen": 592969728 }, { "epoch": 0.18, "learning_rate": 0.0004143129388164494, "loss": 2.385, "theoretical_loss": 3.8445933150708997, "tokens_seen": 593035264 }, { "epoch": 0.18, "objective/train/docs_used": 345976, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.987703561782837, "objective/train/theoretical_loss": 3.844548894014116, "objective/train/tokens_used": 613560800, "theoretical_loss": 3.844548894014116, "tokens_seen": 593100800 }, { "epoch": 0.18, "learning_rate": 0.0004143029087261785, "loss": 2.7096, "theoretical_loss": 3.844548894014116, "tokens_seen": 593100800 }, { "epoch": 0.18, "learning_rate": 0.00041429287863590776, "loss": 2.7678, "theoretical_loss": 3.844504479239645, "tokens_seen": 593166336 }, { "epoch": 0.18, "learning_rate": 0.0004142828485456369, "loss": 2.6941, "theoretical_loss": 3.844460070745905, "tokens_seen": 593231872 }, { "epoch": 0.18, "learning_rate": 0.0004142728184553661, "loss": 2.6844, "theoretical_loss": 3.8444156685313127, "tokens_seen": 593297408 }, { "epoch": 0.18, "learning_rate": 0.0004142627883650953, "loss": 2.9633, "theoretical_loss": 3.8443712725942882, "tokens_seen": 593362944 }, { "epoch": 0.18, "learning_rate": 0.0004142527582748245, "loss": 2.8068, "theoretical_loss": 3.8443268829332498, "tokens_seen": 593428480 }, { "epoch": 0.18, "learning_rate": 0.00041424272818455367, "loss": 2.9583, "theoretical_loss": 3.844282499546619, "tokens_seen": 593494016 }, { "epoch": 0.18, "learning_rate": 0.0004142326980942829, "loss": 2.9592, "theoretical_loss": 3.844238122432814, "tokens_seen": 593559552 }, { "epoch": 0.18, "learning_rate": 0.00041422266800401203, "loss": 2.963, "theoretical_loss": 3.8441937515902573, "tokens_seen": 593625088 }, { "epoch": 0.18, "learning_rate": 0.00041421263791374126, "loss": 2.7788, "theoretical_loss": 3.84414938701737, "tokens_seen": 593690624 }, { "epoch": 0.18, "learning_rate": 0.0004142026078234704, "loss": 2.807, "theoretical_loss": 3.8441050287125744, "tokens_seen": 593756160 }, { "epoch": 0.18, "learning_rate": 0.0004141925777331996, "loss": 2.706, "theoretical_loss": 3.8440606766742933, "tokens_seen": 593821696 }, { "epoch": 0.18, "learning_rate": 0.0004141825476429288, "loss": 2.9724, "theoretical_loss": 3.844016330900949, "tokens_seen": 593887232 }, { "epoch": 0.18, "learning_rate": 0.000414172517552658, "loss": 2.7244, "theoretical_loss": 3.8439719913909673, "tokens_seen": 593952768 }, { "epoch": 0.18, "learning_rate": 0.00041416248746238717, "loss": 3.0772, "theoretical_loss": 3.843927658142771, "tokens_seen": 594018304 }, { "epoch": 0.18, "learning_rate": 0.00041415245737211635, "loss": 2.6304, "theoretical_loss": 3.8438833311547853, "tokens_seen": 594083840 }, { "epoch": 0.18, "learning_rate": 0.00041414242728184553, "loss": 3.0168, "theoretical_loss": 3.843839010425436, "tokens_seen": 594149376 }, { "epoch": 0.18, "learning_rate": 0.00041413239719157477, "loss": 2.8647, "theoretical_loss": 3.8437946959531497, "tokens_seen": 594214912 }, { "epoch": 0.18, "learning_rate": 0.0004141223671013039, "loss": 2.8631, "theoretical_loss": 3.8437503877363524, "tokens_seen": 594280448 }, { "epoch": 0.18, "learning_rate": 0.00041411233701103313, "loss": 2.6276, "theoretical_loss": 3.8437060857734715, "tokens_seen": 594345984 }, { "epoch": 0.18, "learning_rate": 0.0004141023069207623, "loss": 3.0555, "theoretical_loss": 3.843661790062935, "tokens_seen": 594411520 }, { "epoch": 0.18, "learning_rate": 0.0004140922768304915, "loss": 2.9096, "theoretical_loss": 3.8436175006031714, "tokens_seen": 594477056 }, { "epoch": 0.18, "learning_rate": 0.0004140822467402207, "loss": 2.7759, "theoretical_loss": 3.84357321739261, "tokens_seen": 594542592 }, { "epoch": 0.18, "learning_rate": 0.00041407221664994985, "loss": 2.7858, "theoretical_loss": 3.843528940429679, "tokens_seen": 594608128 }, { "epoch": 0.18, "learning_rate": 0.00041406218655967904, "loss": 3.0237, "theoretical_loss": 3.8434846697128098, "tokens_seen": 594673664 }, { "epoch": 0.18, "objective/train/docs_used": 346603, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.868394136428833, "objective/train/theoretical_loss": 3.843440405240433, "objective/train/tokens_used": 615199200, "theoretical_loss": 3.843440405240433, "tokens_seen": 594739200 }, { "epoch": 0.18, "learning_rate": 0.00041405215646940827, "loss": 2.7638, "theoretical_loss": 3.843440405240433, "tokens_seen": 594739200 }, { "epoch": 0.18, "learning_rate": 0.0004140421263791374, "loss": 2.8275, "theoretical_loss": 3.843396147010979, "tokens_seen": 594804736 }, { "epoch": 0.18, "learning_rate": 0.00041403209628886663, "loss": 3.0026, "theoretical_loss": 3.843351895022881, "tokens_seen": 594870272 }, { "epoch": 0.18, "learning_rate": 0.00041402206619859576, "loss": 2.7582, "theoretical_loss": 3.8433076492745695, "tokens_seen": 594935808 }, { "epoch": 0.18, "learning_rate": 0.000414012036108325, "loss": 2.8949, "theoretical_loss": 3.8432634097644787, "tokens_seen": 595001344 }, { "epoch": 0.18, "learning_rate": 0.0004140020060180542, "loss": 2.8077, "theoretical_loss": 3.8432191764910426, "tokens_seen": 595066880 }, { "epoch": 0.18, "learning_rate": 0.00041399197592778336, "loss": 2.7644, "theoretical_loss": 3.8431749494526937, "tokens_seen": 595132416 }, { "epoch": 0.18, "learning_rate": 0.00041398194583751254, "loss": 2.7257, "theoretical_loss": 3.8431307286478678, "tokens_seen": 595197952 }, { "epoch": 0.18, "learning_rate": 0.0004139719157472417, "loss": 2.8675, "theoretical_loss": 3.843086514075, "tokens_seen": 595263488 }, { "epoch": 0.18, "learning_rate": 0.0004139618856569709, "loss": 2.8906, "theoretical_loss": 3.843042305732526, "tokens_seen": 595329024 }, { "epoch": 0.18, "learning_rate": 0.00041395185556670014, "loss": 2.7717, "theoretical_loss": 3.8429981036188816, "tokens_seen": 595394560 }, { "epoch": 0.18, "learning_rate": 0.00041394182547642926, "loss": 2.8548, "theoretical_loss": 3.8429539077325043, "tokens_seen": 595460096 }, { "epoch": 0.18, "learning_rate": 0.0004139317953861585, "loss": 2.9354, "theoretical_loss": 3.8429097180718315, "tokens_seen": 595525632 }, { "epoch": 0.18, "learning_rate": 0.0004139217652958877, "loss": 2.9645, "theoretical_loss": 3.842865534635301, "tokens_seen": 595591168 }, { "epoch": 0.18, "learning_rate": 0.00041391173520561686, "loss": 2.7224, "theoretical_loss": 3.842821357421352, "tokens_seen": 595656704 }, { "epoch": 0.18, "learning_rate": 0.00041390170511534604, "loss": 2.9284, "theoretical_loss": 3.842777186428423, "tokens_seen": 595722240 }, { "epoch": 0.18, "learning_rate": 0.0004138916750250752, "loss": 2.5825, "theoretical_loss": 3.8427330216549533, "tokens_seen": 595787776 }, { "epoch": 0.18, "learning_rate": 0.0004138816449348044, "loss": 2.8972, "theoretical_loss": 3.8426888630993847, "tokens_seen": 595853312 }, { "epoch": 0.18, "learning_rate": 0.00041387161484453364, "loss": 2.4536, "theoretical_loss": 3.8426447107601565, "tokens_seen": 595918848 }, { "epoch": 0.18, "learning_rate": 0.00041386158475426277, "loss": 3.055, "theoretical_loss": 3.8426005646357106, "tokens_seen": 595984384 }, { "epoch": 0.18, "learning_rate": 0.000413851554663992, "loss": 2.8051, "theoretical_loss": 3.8425564247244894, "tokens_seen": 596049920 }, { "epoch": 0.18, "learning_rate": 0.00041384152457372113, "loss": 2.9206, "theoretical_loss": 3.842512291024935, "tokens_seen": 596115456 }, { "epoch": 0.18, "learning_rate": 0.00041383149448345036, "loss": 2.9248, "theoretical_loss": 3.842468163535491, "tokens_seen": 596180992 }, { "epoch": 0.18, "learning_rate": 0.00041382146439317954, "loss": 2.8616, "theoretical_loss": 3.842424042254601, "tokens_seen": 596246528 }, { "epoch": 0.18, "learning_rate": 0.0004138114343029087, "loss": 2.8664, "theoretical_loss": 3.8423799271807084, "tokens_seen": 596312064 }, { "epoch": 0.18, "objective/train/docs_used": 348035, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6039180755615234, "objective/train/theoretical_loss": 3.8423358183122582, "objective/train/tokens_used": 616837600, "theoretical_loss": 3.8423358183122582, "tokens_seen": 596377600 }, { "epoch": 0.18, "learning_rate": 0.0004138014042126379, "loss": 2.8114, "theoretical_loss": 3.8423358183122582, "tokens_seen": 596377600 }, { "epoch": 0.18, "learning_rate": 0.0004137913741223671, "loss": 2.8918, "theoretical_loss": 3.8422917156476966, "tokens_seen": 596443136 }, { "epoch": 0.18, "learning_rate": 0.00041378134403209627, "loss": 2.9722, "theoretical_loss": 3.8422476191854686, "tokens_seen": 596508672 }, { "epoch": 0.18, "learning_rate": 0.0004137713139418255, "loss": 2.757, "theoretical_loss": 3.8422035289240215, "tokens_seen": 596574208 }, { "epoch": 0.18, "learning_rate": 0.00041376128385155463, "loss": 2.6994, "theoretical_loss": 3.8421594448618013, "tokens_seen": 596639744 }, { "epoch": 0.18, "learning_rate": 0.00041375125376128387, "loss": 2.7965, "theoretical_loss": 3.8421153669972563, "tokens_seen": 596705280 }, { "epoch": 0.18, "learning_rate": 0.00041374122367101305, "loss": 2.6647, "theoretical_loss": 3.842071295328835, "tokens_seen": 596770816 }, { "epoch": 0.18, "learning_rate": 0.00041373119358074223, "loss": 2.7225, "theoretical_loss": 3.8420272298549847, "tokens_seen": 596836352 }, { "epoch": 0.18, "learning_rate": 0.0004137211634904714, "loss": 2.8572, "theoretical_loss": 3.8419831705741556, "tokens_seen": 596901888 }, { "epoch": 0.18, "learning_rate": 0.0004137111334002006, "loss": 2.7221, "theoretical_loss": 3.841939117484798, "tokens_seen": 596967424 }, { "epoch": 0.18, "learning_rate": 0.00041370110330992977, "loss": 2.7622, "theoretical_loss": 3.841895070585361, "tokens_seen": 597032960 }, { "epoch": 0.18, "learning_rate": 0.000413691073219659, "loss": 2.8918, "theoretical_loss": 3.8418510298742965, "tokens_seen": 597098496 }, { "epoch": 0.18, "learning_rate": 0.00041368104312938813, "loss": 2.869, "theoretical_loss": 3.8418069953500558, "tokens_seen": 597164032 }, { "epoch": 0.18, "learning_rate": 0.00041367101303911737, "loss": 2.8955, "theoretical_loss": 3.8417629670110904, "tokens_seen": 597229568 }, { "epoch": 0.18, "learning_rate": 0.00041366098294884655, "loss": 2.7212, "theoretical_loss": 3.841718944855854, "tokens_seen": 597295104 }, { "epoch": 0.18, "learning_rate": 0.00041365095285857573, "loss": 2.7575, "theoretical_loss": 3.8416749288827985, "tokens_seen": 597360640 }, { "epoch": 0.18, "learning_rate": 0.00041364092276830497, "loss": 3.0717, "theoretical_loss": 3.8416309190903783, "tokens_seen": 597426176 }, { "epoch": 0.18, "learning_rate": 0.0004136308926780341, "loss": 2.8992, "theoretical_loss": 3.841586915477048, "tokens_seen": 597491712 }, { "epoch": 0.18, "learning_rate": 0.00041362086258776333, "loss": 2.7046, "theoretical_loss": 3.8415429180412617, "tokens_seen": 597557248 }, { "epoch": 0.18, "learning_rate": 0.0004136108324974925, "loss": 2.8873, "theoretical_loss": 3.841498926781475, "tokens_seen": 597622784 }, { "epoch": 0.18, "learning_rate": 0.0004136008024072217, "loss": 2.8448, "theoretical_loss": 3.8414549416961443, "tokens_seen": 597688320 }, { "epoch": 0.18, "learning_rate": 0.0004135907723169509, "loss": 2.795, "theoretical_loss": 3.8414109627837254, "tokens_seen": 597753856 }, { "epoch": 0.18, "learning_rate": 0.00041358074222668005, "loss": 2.8589, "theoretical_loss": 3.841366990042676, "tokens_seen": 597819392 }, { "epoch": 0.18, "learning_rate": 0.00041357071213640924, "loss": 2.636, "theoretical_loss": 3.8413230234714533, "tokens_seen": 597884928 }, { "epoch": 0.18, "learning_rate": 0.00041356068204613847, "loss": 2.8705, "theoretical_loss": 3.8412790630685154, "tokens_seen": 597950464 }, { "epoch": 0.18, "objective/train/docs_used": 348629, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.116250514984131, "objective/train/theoretical_loss": 3.841235108832321, "objective/train/tokens_used": 618476000, "theoretical_loss": 3.841235108832321, "tokens_seen": 598016000 }, { "epoch": 0.18, "learning_rate": 0.0004135506519558676, "loss": 3.0489, "theoretical_loss": 3.841235108832321, "tokens_seen": 598016000 }, { "epoch": 0.18, "learning_rate": 0.00041354062186559683, "loss": 2.7793, "theoretical_loss": 3.84119116076133, "tokens_seen": 598081536 }, { "epoch": 0.18, "learning_rate": 0.00041353059177532596, "loss": 2.9613, "theoretical_loss": 3.841147218854002, "tokens_seen": 598147072 }, { "epoch": 0.18, "learning_rate": 0.0004135205616850552, "loss": 2.6289, "theoretical_loss": 3.8411032831087963, "tokens_seen": 598212608 }, { "epoch": 0.18, "learning_rate": 0.0004135105315947844, "loss": 2.8951, "theoretical_loss": 3.841059353524175, "tokens_seen": 598278144 }, { "epoch": 0.18, "learning_rate": 0.00041350050150451356, "loss": 2.8064, "theoretical_loss": 3.841015430098599, "tokens_seen": 598343680 }, { "epoch": 0.18, "learning_rate": 0.00041349047141424274, "loss": 2.6171, "theoretical_loss": 3.8409715128305306, "tokens_seen": 598409216 }, { "epoch": 0.18, "learning_rate": 0.0004134804413239719, "loss": 2.6367, "theoretical_loss": 3.8409276017184326, "tokens_seen": 598474752 }, { "epoch": 0.18, "learning_rate": 0.0004134704112337011, "loss": 2.765, "theoretical_loss": 3.840883696760768, "tokens_seen": 598540288 }, { "epoch": 0.18, "learning_rate": 0.00041346038114343034, "loss": 2.6495, "theoretical_loss": 3.840839797956, "tokens_seen": 598605824 }, { "epoch": 0.18, "learning_rate": 0.00041345035105315946, "loss": 2.6851, "theoretical_loss": 3.8407959053025937, "tokens_seen": 598671360 }, { "epoch": 0.18, "learning_rate": 0.0004134403209628887, "loss": 2.8062, "theoretical_loss": 3.8407520187990127, "tokens_seen": 598736896 }, { "epoch": 0.18, "learning_rate": 0.0004134302908726179, "loss": 2.7218, "theoretical_loss": 3.8407081384437234, "tokens_seen": 598802432 }, { "epoch": 0.18, "learning_rate": 0.00041342026078234706, "loss": 2.7344, "theoretical_loss": 3.8406642642351914, "tokens_seen": 598867968 }, { "epoch": 0.18, "learning_rate": 0.00041341023069207624, "loss": 2.5827, "theoretical_loss": 3.840620396171883, "tokens_seen": 598933504 }, { "epoch": 0.18, "learning_rate": 0.0004134002006018054, "loss": 3.0092, "theoretical_loss": 3.840576534252265, "tokens_seen": 598999040 }, { "epoch": 0.18, "learning_rate": 0.0004133901705115346, "loss": 2.8179, "theoretical_loss": 3.8405326784748057, "tokens_seen": 599064576 }, { "epoch": 0.18, "learning_rate": 0.00041338014042126384, "loss": 2.8732, "theoretical_loss": 3.840488828837973, "tokens_seen": 599130112 }, { "epoch": 0.18, "learning_rate": 0.00041337011033099297, "loss": 2.9042, "theoretical_loss": 3.840444985340235, "tokens_seen": 599195648 }, { "epoch": 0.18, "learning_rate": 0.0004133600802407222, "loss": 2.9798, "theoretical_loss": 3.840401147980061, "tokens_seen": 599261184 }, { "epoch": 0.18, "learning_rate": 0.00041335005015045133, "loss": 2.5213, "theoretical_loss": 3.840357316755921, "tokens_seen": 599326720 }, { "epoch": 0.18, "learning_rate": 0.00041334002006018056, "loss": 3.0255, "theoretical_loss": 3.8403134916662847, "tokens_seen": 599392256 }, { "epoch": 0.18, "learning_rate": 0.00041332998996990975, "loss": 2.6682, "theoretical_loss": 3.840269672709624, "tokens_seen": 599457792 }, { "epoch": 0.18, "learning_rate": 0.0004133199598796389, "loss": 2.9696, "theoretical_loss": 3.8402258598844097, "tokens_seen": 599523328 }, { "epoch": 0.18, "learning_rate": 0.0004133099297893681, "loss": 2.8461, "theoretical_loss": 3.8401820531891135, "tokens_seen": 599588864 }, { "epoch": 0.18, "objective/train/docs_used": 349996, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6275978088378906, "objective/train/theoretical_loss": 3.840138252622208, "objective/train/tokens_used": 620114400, "theoretical_loss": 3.840138252622208, "tokens_seen": 599654400 }, { "epoch": 0.18, "learning_rate": 0.0004132998996990973, "loss": 2.7282, "theoretical_loss": 3.840138252622208, "tokens_seen": 599654400 }, { "epoch": 0.18, "learning_rate": 0.00041328986960882647, "loss": 2.7441, "theoretical_loss": 3.8400944581821674, "tokens_seen": 599719936 }, { "epoch": 0.18, "learning_rate": 0.0004132798395185557, "loss": 3.025, "theoretical_loss": 3.8400506698674635, "tokens_seen": 599785472 }, { "epoch": 0.18, "learning_rate": 0.00041326980942828483, "loss": 2.7853, "theoretical_loss": 3.840006887676571, "tokens_seen": 599851008 }, { "epoch": 0.18, "learning_rate": 0.00041325977933801407, "loss": 2.9446, "theoretical_loss": 3.839963111607965, "tokens_seen": 599916544 }, { "epoch": 0.18, "learning_rate": 0.00041324974924774325, "loss": 2.9363, "theoretical_loss": 3.8399193416601207, "tokens_seen": 599982080 }, { "epoch": 0.18, "learning_rate": 0.00041323971915747243, "loss": 2.6701, "theoretical_loss": 3.8398755778315135, "tokens_seen": 600047616 }, { "epoch": 0.18, "learning_rate": 0.0004132296890672016, "loss": 2.695, "theoretical_loss": 3.8398318201206196, "tokens_seen": 600113152 }, { "epoch": 0.18, "learning_rate": 0.0004132196589769308, "loss": 2.9478, "theoretical_loss": 3.8397880685259165, "tokens_seen": 600178688 }, { "epoch": 0.18, "learning_rate": 0.00041320962888665997, "loss": 2.9884, "theoretical_loss": 3.839744323045881, "tokens_seen": 600244224 }, { "epoch": 0.18, "learning_rate": 0.0004131995987963892, "loss": 2.556, "theoretical_loss": 3.839700583678991, "tokens_seen": 600309760 }, { "epoch": 0.18, "learning_rate": 0.00041318956870611833, "loss": 2.7496, "theoretical_loss": 3.839656850423726, "tokens_seen": 600375296 }, { "epoch": 0.18, "learning_rate": 0.00041317953861584757, "loss": 2.791, "theoretical_loss": 3.8396131232785637, "tokens_seen": 600440832 }, { "epoch": 0.18, "learning_rate": 0.0004131695085255767, "loss": 2.6448, "theoretical_loss": 3.8395694022419846, "tokens_seen": 600506368 }, { "epoch": 0.18, "learning_rate": 0.00041315947843530593, "loss": 3.0441, "theoretical_loss": 3.8395256873124683, "tokens_seen": 600571904 }, { "epoch": 0.18, "learning_rate": 0.0004131494483450351, "loss": 2.9385, "theoretical_loss": 3.8394819784884957, "tokens_seen": 600637440 }, { "epoch": 0.18, "learning_rate": 0.0004131394182547643, "loss": 2.6955, "theoretical_loss": 3.8394382757685483, "tokens_seen": 600702976 }, { "epoch": 0.18, "learning_rate": 0.0004131293881644935, "loss": 2.7299, "theoretical_loss": 3.839394579151107, "tokens_seen": 600768512 }, { "epoch": 0.18, "learning_rate": 0.0004131193580742227, "loss": 2.7899, "theoretical_loss": 3.8393508886346552, "tokens_seen": 600834048 }, { "epoch": 0.18, "learning_rate": 0.00041310932798395184, "loss": 2.7416, "theoretical_loss": 3.8393072042176755, "tokens_seen": 600899584 }, { "epoch": 0.18, "learning_rate": 0.0004130992978936811, "loss": 2.7468, "theoretical_loss": 3.83926352589865, "tokens_seen": 600965120 }, { "epoch": 0.18, "learning_rate": 0.0004130892678034102, "loss": 2.5742, "theoretical_loss": 3.839219853676064, "tokens_seen": 601030656 }, { "epoch": 0.18, "learning_rate": 0.00041307923771313944, "loss": 2.8156, "theoretical_loss": 3.839176187548402, "tokens_seen": 601096192 }, { "epoch": 0.18, "learning_rate": 0.0004130692076228686, "loss": 2.8401, "theoretical_loss": 3.839132527514148, "tokens_seen": 601161728 }, { "epoch": 0.18, "learning_rate": 0.0004130591775325978, "loss": 2.7502, "theoretical_loss": 3.8390888735717885, "tokens_seen": 601227264 }, { "epoch": 0.18, "objective/train/docs_used": 350686, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7006852626800537, "objective/train/theoretical_loss": 3.8390452257198096, "objective/train/tokens_used": 621752800, "theoretical_loss": 3.8390452257198096, "tokens_seen": 601292800 }, { "epoch": 0.18, "learning_rate": 0.000413049147442327, "loss": 2.7745, "theoretical_loss": 3.8390452257198096, "tokens_seen": 601292800 }, { "epoch": 0.18, "learning_rate": 0.00041303911735205616, "loss": 3.0797, "theoretical_loss": 3.839001583956697, "tokens_seen": 601358336 }, { "epoch": 0.18, "learning_rate": 0.00041302908726178534, "loss": 3.1748, "theoretical_loss": 3.838957948280939, "tokens_seen": 601423872 }, { "epoch": 0.18, "learning_rate": 0.0004130190571715146, "loss": 2.9787, "theoretical_loss": 3.8389143186910224, "tokens_seen": 601489408 }, { "epoch": 0.18, "learning_rate": 0.0004130090270812437, "loss": 3.0155, "theoretical_loss": 3.8388706951854354, "tokens_seen": 601554944 }, { "epoch": 0.18, "learning_rate": 0.00041299899699097294, "loss": 2.7272, "theoretical_loss": 3.838827077762668, "tokens_seen": 601620480 }, { "epoch": 0.18, "learning_rate": 0.00041298896690070207, "loss": 2.6408, "theoretical_loss": 3.8387834664212086, "tokens_seen": 601686016 }, { "epoch": 0.18, "learning_rate": 0.0004129789368104313, "loss": 2.8119, "theoretical_loss": 3.8387398611595476, "tokens_seen": 601751552 }, { "epoch": 0.18, "learning_rate": 0.0004129689067201605, "loss": 2.8147, "theoretical_loss": 3.8386962619761746, "tokens_seen": 601817088 }, { "epoch": 0.18, "learning_rate": 0.00041295887662988966, "loss": 2.6513, "theoretical_loss": 3.838652668869581, "tokens_seen": 601882624 }, { "epoch": 0.18, "learning_rate": 0.00041294884653961884, "loss": 2.7751, "theoretical_loss": 3.8386090818382588, "tokens_seen": 601948160 }, { "epoch": 0.18, "learning_rate": 0.0004129388164493481, "loss": 2.9408, "theoretical_loss": 3.838565500880699, "tokens_seen": 602013696 }, { "epoch": 0.18, "learning_rate": 0.0004129287863590772, "loss": 2.5206, "theoretical_loss": 3.838521925995395, "tokens_seen": 602079232 }, { "epoch": 0.18, "learning_rate": 0.00041291875626880644, "loss": 2.9071, "theoretical_loss": 3.83847835718084, "tokens_seen": 602144768 }, { "epoch": 0.18, "learning_rate": 0.0004129087261785356, "loss": 2.836, "theoretical_loss": 3.838434794435527, "tokens_seen": 602210304 }, { "epoch": 0.18, "learning_rate": 0.0004128986960882648, "loss": 2.432, "theoretical_loss": 3.8383912377579508, "tokens_seen": 602275840 }, { "epoch": 0.18, "learning_rate": 0.00041288866599799404, "loss": 2.7358, "theoretical_loss": 3.838347687146605, "tokens_seen": 602341376 }, { "epoch": 0.18, "learning_rate": 0.00041287863590772317, "loss": 2.8274, "theoretical_loss": 3.8383041425999864, "tokens_seen": 602406912 }, { "epoch": 0.18, "learning_rate": 0.0004128686058174524, "loss": 2.7196, "theoretical_loss": 3.83826060411659, "tokens_seen": 602472448 }, { "epoch": 0.18, "learning_rate": 0.00041285857572718153, "loss": 2.6672, "theoretical_loss": 3.8382170716949124, "tokens_seen": 602537984 }, { "epoch": 0.18, "learning_rate": 0.00041284854563691076, "loss": 2.8981, "theoretical_loss": 3.83817354533345, "tokens_seen": 602603520 }, { "epoch": 0.18, "learning_rate": 0.00041283851554663995, "loss": 2.6895, "theoretical_loss": 3.8381300250307007, "tokens_seen": 602669056 }, { "epoch": 0.18, "learning_rate": 0.0004128284854563691, "loss": 2.525, "theoretical_loss": 3.8380865107851623, "tokens_seen": 602734592 }, { "epoch": 0.18, "learning_rate": 0.0004128184553660983, "loss": 2.618, "theoretical_loss": 3.8380430025953327, "tokens_seen": 602800128 }, { "epoch": 0.18, "learning_rate": 0.0004128084252758275, "loss": 2.5809, "theoretical_loss": 3.837999500459712, "tokens_seen": 602865664 }, { "epoch": 0.18, "objective/train/docs_used": 352047, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8010265827178955, "objective/train/theoretical_loss": 3.837956004376799, "objective/train/tokens_used": 623391200, "theoretical_loss": 3.837956004376799, "tokens_seen": 602931200 }, { "epoch": 0.18, "learning_rate": 0.00041279839518555667, "loss": 2.56, "theoretical_loss": 3.837956004376799, "tokens_seen": 602931200 }, { "epoch": 0.18, "learning_rate": 0.0004127883650952859, "loss": 2.8099, "theoretical_loss": 3.837912514345094, "tokens_seen": 602996736 }, { "epoch": 0.18, "learning_rate": 0.00041277833500501503, "loss": 3.0077, "theoretical_loss": 3.8378690303630973, "tokens_seen": 603062272 }, { "epoch": 0.18, "learning_rate": 0.00041276830491474427, "loss": 2.8896, "theoretical_loss": 3.837825552429311, "tokens_seen": 603127808 }, { "epoch": 0.18, "learning_rate": 0.00041275827482447345, "loss": 2.7193, "theoretical_loss": 3.837782080542236, "tokens_seen": 603193344 }, { "epoch": 0.18, "learning_rate": 0.00041274824473420263, "loss": 2.8501, "theoretical_loss": 3.8377386147003745, "tokens_seen": 603258880 }, { "epoch": 0.18, "learning_rate": 0.0004127382146439318, "loss": 2.6467, "theoretical_loss": 3.837695154902229, "tokens_seen": 603324416 }, { "epoch": 0.18, "learning_rate": 0.000412728184553661, "loss": 2.787, "theoretical_loss": 3.837651701146304, "tokens_seen": 603389952 }, { "epoch": 0.18, "learning_rate": 0.00041271815446339017, "loss": 2.7283, "theoretical_loss": 3.8376082534311022, "tokens_seen": 603455488 }, { "epoch": 0.18, "learning_rate": 0.0004127081243731194, "loss": 2.8614, "theoretical_loss": 3.8375648117551284, "tokens_seen": 603521024 }, { "epoch": 0.18, "learning_rate": 0.00041269809428284854, "loss": 2.7903, "theoretical_loss": 3.8375213761168867, "tokens_seen": 603586560 }, { "epoch": 0.18, "learning_rate": 0.00041268806419257777, "loss": 2.729, "theoretical_loss": 3.837477946514884, "tokens_seen": 603652096 }, { "epoch": 0.18, "learning_rate": 0.0004126780341023069, "loss": 2.9156, "theoretical_loss": 3.8374345229476248, "tokens_seen": 603717632 }, { "epoch": 0.18, "learning_rate": 0.00041266800401203613, "loss": 2.8972, "theoretical_loss": 3.8373911054136167, "tokens_seen": 603783168 }, { "epoch": 0.18, "learning_rate": 0.0004126579739217653, "loss": 2.6882, "theoretical_loss": 3.837347693911366, "tokens_seen": 603848704 }, { "epoch": 0.18, "learning_rate": 0.0004126479438314945, "loss": 2.9854, "theoretical_loss": 3.8373042884393813, "tokens_seen": 603914240 }, { "epoch": 0.18, "learning_rate": 0.0004126379137412237, "loss": 2.725, "theoretical_loss": 3.837260888996169, "tokens_seen": 603979776 }, { "epoch": 0.18, "learning_rate": 0.0004126278836509529, "loss": 2.8998, "theoretical_loss": 3.837217495580239, "tokens_seen": 604045312 }, { "epoch": 0.18, "learning_rate": 0.00041261785356068204, "loss": 2.9865, "theoretical_loss": 3.8371741081900996, "tokens_seen": 604110848 }, { "epoch": 0.18, "learning_rate": 0.0004126078234704113, "loss": 3.0078, "theoretical_loss": 3.8371307268242614, "tokens_seen": 604176384 }, { "epoch": 0.18, "learning_rate": 0.0004125977933801404, "loss": 2.9132, "theoretical_loss": 3.837087351481234, "tokens_seen": 604241920 }, { "epoch": 0.18, "learning_rate": 0.00041258776328986964, "loss": 2.8328, "theoretical_loss": 3.8370439821595284, "tokens_seen": 604307456 }, { "epoch": 0.18, "learning_rate": 0.0004125777331995988, "loss": 2.5516, "theoretical_loss": 3.8370006188576555, "tokens_seen": 604372992 }, { "epoch": 0.18, "learning_rate": 0.000412567703109328, "loss": 2.9134, "theoretical_loss": 3.8369572615741268, "tokens_seen": 604438528 }, { "epoch": 0.18, "learning_rate": 0.0004125576730190572, "loss": 2.9092, "theoretical_loss": 3.836913910307456, "tokens_seen": 604504064 }, { "epoch": 0.18, "objective/train/docs_used": 352465, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7846691608428955, "objective/train/theoretical_loss": 3.836870565056154, "objective/train/tokens_used": 625029600, "theoretical_loss": 3.836870565056154, "tokens_seen": 604569600 }, { "epoch": 0.18, "learning_rate": 0.00041254764292878636, "loss": 2.7988, "theoretical_loss": 3.836870565056154, "tokens_seen": 604569600 }, { "epoch": 0.18, "learning_rate": 0.00041253761283851554, "loss": 2.7453, "theoretical_loss": 3.8368272258187366, "tokens_seen": 604635136 }, { "epoch": 0.18, "learning_rate": 0.0004125275827482448, "loss": 2.9183, "theoretical_loss": 3.836783892593715, "tokens_seen": 604700672 }, { "epoch": 0.18, "learning_rate": 0.0004125175526579739, "loss": 2.8628, "theoretical_loss": 3.836740565379606, "tokens_seen": 604766208 }, { "epoch": 0.18, "learning_rate": 0.00041250752256770314, "loss": 2.7834, "theoretical_loss": 3.836697244174923, "tokens_seen": 604831744 }, { "epoch": 0.18, "learning_rate": 0.00041249749247743227, "loss": 2.6896, "theoretical_loss": 3.836653928978182, "tokens_seen": 604897280 }, { "epoch": 0.18, "learning_rate": 0.0004124874623871615, "loss": 2.9959, "theoretical_loss": 3.8366106197878995, "tokens_seen": 604962816 }, { "epoch": 0.18, "learning_rate": 0.0004124774322968907, "loss": 2.9265, "theoretical_loss": 3.8365673166025913, "tokens_seen": 605028352 }, { "epoch": 0.18, "learning_rate": 0.00041246740220661986, "loss": 2.7492, "theoretical_loss": 3.8365240194207746, "tokens_seen": 605093888 }, { "epoch": 0.18, "learning_rate": 0.00041245737211634904, "loss": 2.8017, "theoretical_loss": 3.8364807282409674, "tokens_seen": 605159424 }, { "epoch": 0.18, "learning_rate": 0.0004124473420260783, "loss": 2.8394, "theoretical_loss": 3.836437443061688, "tokens_seen": 605224960 }, { "epoch": 0.18, "learning_rate": 0.0004124373119358074, "loss": 2.9353, "theoretical_loss": 3.8363941638814536, "tokens_seen": 605290496 }, { "epoch": 0.18, "learning_rate": 0.00041242728184553664, "loss": 2.5652, "theoretical_loss": 3.8363508906987853, "tokens_seen": 605356032 }, { "epoch": 0.18, "learning_rate": 0.00041241725175526577, "loss": 2.7886, "theoretical_loss": 3.8363076235122016, "tokens_seen": 605421568 }, { "epoch": 0.18, "learning_rate": 0.000412407221664995, "loss": 2.6518, "theoretical_loss": 3.8362643623202226, "tokens_seen": 605487104 }, { "epoch": 0.18, "learning_rate": 0.0004123971915747242, "loss": 2.9449, "theoretical_loss": 3.83622110712137, "tokens_seen": 605552640 }, { "epoch": 0.18, "learning_rate": 0.00041238716148445337, "loss": 2.871, "theoretical_loss": 3.8361778579141648, "tokens_seen": 605618176 }, { "epoch": 0.18, "learning_rate": 0.00041237713139418255, "loss": 2.8066, "theoretical_loss": 3.8361346146971282, "tokens_seen": 605683712 }, { "epoch": 0.18, "learning_rate": 0.00041236710130391173, "loss": 3.2092, "theoretical_loss": 3.836091377468783, "tokens_seen": 605749248 }, { "epoch": 0.18, "learning_rate": 0.0004123570712136409, "loss": 2.7747, "theoretical_loss": 3.836048146227652, "tokens_seen": 605814784 }, { "epoch": 0.18, "learning_rate": 0.00041234704112337015, "loss": 2.6126, "theoretical_loss": 3.8360049209722584, "tokens_seen": 605880320 }, { "epoch": 0.18, "learning_rate": 0.00041233701103309927, "loss": 3.015, "theoretical_loss": 3.8359617017011267, "tokens_seen": 605945856 }, { "epoch": 0.18, "learning_rate": 0.0004123269809428285, "loss": 2.5441, "theoretical_loss": 3.835918488412781, "tokens_seen": 606011392 }, { "epoch": 0.18, "learning_rate": 0.00041231695085255763, "loss": 2.5169, "theoretical_loss": 3.8358752811057455, "tokens_seen": 606076928 }, { "epoch": 0.18, "learning_rate": 0.00041230692076228687, "loss": 2.8302, "theoretical_loss": 3.8358320797785463, "tokens_seen": 606142464 }, { "epoch": 0.18, "objective/train/docs_used": 353634, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.123135566711426, "objective/train/theoretical_loss": 3.8357888844297094, "objective/train/tokens_used": 626668000, "theoretical_loss": 3.8357888844297094, "tokens_seen": 606208000 }, { "epoch": 0.18, "learning_rate": 0.00041229689067201605, "loss": 2.9484, "theoretical_loss": 3.8357888844297094, "tokens_seen": 606208000 }, { "epoch": 0.18, "learning_rate": 0.00041228686058174523, "loss": 2.9537, "theoretical_loss": 3.8357456950577618, "tokens_seen": 606273536 }, { "epoch": 0.18, "learning_rate": 0.0004122768304914744, "loss": 2.6897, "theoretical_loss": 3.8357025116612293, "tokens_seen": 606339072 }, { "epoch": 0.18, "learning_rate": 0.00041226680040120365, "loss": 2.813, "theoretical_loss": 3.8356593342386405, "tokens_seen": 606404608 }, { "epoch": 0.18, "learning_rate": 0.0004122567703109328, "loss": 2.8278, "theoretical_loss": 3.8356161627885235, "tokens_seen": 606470144 }, { "epoch": 0.18, "learning_rate": 0.000412246740220662, "loss": 3.0245, "theoretical_loss": 3.835572997309406, "tokens_seen": 606535680 }, { "epoch": 0.18, "learning_rate": 0.00041223671013039114, "loss": 2.8529, "theoretical_loss": 3.835529837799818, "tokens_seen": 606601216 }, { "epoch": 0.18, "learning_rate": 0.0004122266800401204, "loss": 2.6315, "theoretical_loss": 3.8354866842582895, "tokens_seen": 606666752 }, { "epoch": 0.18, "learning_rate": 0.00041221664994984955, "loss": 2.932, "theoretical_loss": 3.835443536683349, "tokens_seen": 606732288 }, { "epoch": 0.18, "learning_rate": 0.00041220661985957874, "loss": 2.622, "theoretical_loss": 3.835400395073529, "tokens_seen": 606797824 }, { "epoch": 0.18, "learning_rate": 0.0004121965897693079, "loss": 2.9286, "theoretical_loss": 3.8353572594273597, "tokens_seen": 606863360 }, { "epoch": 0.18, "learning_rate": 0.0004121865596790371, "loss": 2.9135, "theoretical_loss": 3.835314129743373, "tokens_seen": 606928896 }, { "epoch": 0.18, "learning_rate": 0.0004121765295887663, "loss": 2.8349, "theoretical_loss": 3.835271006020101, "tokens_seen": 606994432 }, { "epoch": 0.18, "learning_rate": 0.0004121664994984955, "loss": 2.8695, "theoretical_loss": 3.8352278882560764, "tokens_seen": 607059968 }, { "epoch": 0.18, "learning_rate": 0.0004121564694082247, "loss": 2.9327, "theoretical_loss": 3.8351847764498332, "tokens_seen": 607125504 }, { "epoch": 0.18, "learning_rate": 0.0004121464393179539, "loss": 2.7951, "theoretical_loss": 3.8351416705999055, "tokens_seen": 607191040 }, { "epoch": 0.18, "learning_rate": 0.0004121364092276831, "loss": 2.7969, "theoretical_loss": 3.835098570704826, "tokens_seen": 607256576 }, { "epoch": 0.18, "learning_rate": 0.00041212637913741224, "loss": 2.9292, "theoretical_loss": 3.8350554767631304, "tokens_seen": 607322112 }, { "epoch": 0.18, "learning_rate": 0.0004121163490471415, "loss": 2.9228, "theoretical_loss": 3.8350123887733543, "tokens_seen": 607387648 }, { "epoch": 0.18, "learning_rate": 0.0004121063189568706, "loss": 2.6295, "theoretical_loss": 3.8349693067340334, "tokens_seen": 607453184 }, { "epoch": 0.18, "learning_rate": 0.00041209628886659984, "loss": 3.0163, "theoretical_loss": 3.8349262306437044, "tokens_seen": 607518720 }, { "epoch": 0.18, "learning_rate": 0.000412086258776329, "loss": 3.0652, "theoretical_loss": 3.8348831605009033, "tokens_seen": 607584256 }, { "epoch": 0.18, "learning_rate": 0.0004120762286860582, "loss": 2.6424, "theoretical_loss": 3.8348400963041684, "tokens_seen": 607649792 }, { "epoch": 0.18, "learning_rate": 0.0004120661985957874, "loss": 2.9793, "theoretical_loss": 3.8347970380520375, "tokens_seen": 607715328 }, { "epoch": 0.18, "learning_rate": 0.00041205616850551656, "loss": 2.9131, "theoretical_loss": 3.834753985743049, "tokens_seen": 607780864 }, { "epoch": 0.18, "objective/train/docs_used": 354263, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.821697950363159, "objective/train/theoretical_loss": 3.834710939375742, "objective/train/tokens_used": 628306400, "theoretical_loss": 3.834710939375742, "tokens_seen": 607846400 }, { "epoch": 0.18, "learning_rate": 0.00041204613841524574, "loss": 2.8469, "theoretical_loss": 3.834710939375742, "tokens_seen": 607846400 }, { "epoch": 0.18, "learning_rate": 0.000412036108324975, "loss": 2.5941, "theoretical_loss": 3.8346678989486556, "tokens_seen": 607911936 }, { "epoch": 0.18, "learning_rate": 0.0004120260782347041, "loss": 2.874, "theoretical_loss": 3.83462486446033, "tokens_seen": 607977472 }, { "epoch": 0.18, "learning_rate": 0.00041201604814443334, "loss": 2.7832, "theoretical_loss": 3.8345818359093062, "tokens_seen": 608043008 }, { "epoch": 0.18, "learning_rate": 0.00041200601805416247, "loss": 2.8339, "theoretical_loss": 3.8345388132941243, "tokens_seen": 608108544 }, { "epoch": 0.18, "learning_rate": 0.0004119959879638917, "loss": 2.8034, "theoretical_loss": 3.834495796613327, "tokens_seen": 608174080 }, { "epoch": 0.18, "learning_rate": 0.0004119859578736209, "loss": 3.2167, "theoretical_loss": 3.834452785865455, "tokens_seen": 608239616 }, { "epoch": 0.18, "learning_rate": 0.00041197592778335006, "loss": 2.6944, "theoretical_loss": 3.834409781049053, "tokens_seen": 608305152 }, { "epoch": 0.18, "learning_rate": 0.00041196589769307924, "loss": 2.9677, "theoretical_loss": 3.834366782162662, "tokens_seen": 608370688 }, { "epoch": 0.18, "learning_rate": 0.0004119558676028085, "loss": 2.8102, "theoretical_loss": 3.8343237892048263, "tokens_seen": 608436224 }, { "epoch": 0.18, "learning_rate": 0.0004119458375125376, "loss": 2.806, "theoretical_loss": 3.834280802174091, "tokens_seen": 608501760 }, { "epoch": 0.18, "learning_rate": 0.00041193580742226684, "loss": 2.7141, "theoretical_loss": 3.8342378210689994, "tokens_seen": 608567296 }, { "epoch": 0.18, "learning_rate": 0.00041192577733199597, "loss": 2.4579, "theoretical_loss": 3.834194845888097, "tokens_seen": 608632832 }, { "epoch": 0.18, "learning_rate": 0.0004119157472417252, "loss": 2.7697, "theoretical_loss": 3.83415187662993, "tokens_seen": 608698368 }, { "epoch": 0.18, "learning_rate": 0.0004119057171514544, "loss": 2.8852, "theoretical_loss": 3.8341089132930444, "tokens_seen": 608763904 }, { "epoch": 0.18, "learning_rate": 0.00041189568706118357, "loss": 2.9821, "theoretical_loss": 3.834065955875987, "tokens_seen": 608829440 }, { "epoch": 0.18, "learning_rate": 0.00041188565697091275, "loss": 2.873, "theoretical_loss": 3.8340230043773045, "tokens_seen": 608894976 }, { "epoch": 0.18, "learning_rate": 0.00041187562688064193, "loss": 2.8123, "theoretical_loss": 3.833980058795545, "tokens_seen": 608960512 }, { "epoch": 0.18, "learning_rate": 0.0004118655967903711, "loss": 2.8689, "theoretical_loss": 3.8339371191292573, "tokens_seen": 609026048 }, { "epoch": 0.18, "learning_rate": 0.00041185556670010035, "loss": 2.7258, "theoretical_loss": 3.8338941853769892, "tokens_seen": 609091584 }, { "epoch": 0.18, "learning_rate": 0.00041184553660982947, "loss": 3.06, "theoretical_loss": 3.8338512575372903, "tokens_seen": 609157120 }, { "epoch": 0.18, "learning_rate": 0.0004118355065195587, "loss": 2.7505, "theoretical_loss": 3.8338083356087105, "tokens_seen": 609222656 }, { "epoch": 0.18, "learning_rate": 0.00041182547642928783, "loss": 2.8264, "theoretical_loss": 3.8337654195898, "tokens_seen": 609288192 }, { "epoch": 0.18, "learning_rate": 0.00041181544633901707, "loss": 2.7775, "theoretical_loss": 3.8337225094791103, "tokens_seen": 609353728 }, { "epoch": 0.18, "learning_rate": 0.00041180541624874625, "loss": 2.9664, "theoretical_loss": 3.8336796052751914, "tokens_seen": 609419264 }, { "epoch": 0.18, "objective/train/docs_used": 355721, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.350303888320923, "objective/train/theoretical_loss": 3.8336367069765958, "objective/train/tokens_used": 629944800, "theoretical_loss": 3.8336367069765958, "tokens_seen": 609484800 }, { "epoch": 0.18, "learning_rate": 0.00041179538615847543, "loss": 2.6211, "theoretical_loss": 3.8336367069765958, "tokens_seen": 609484800 }, { "epoch": 0.18, "learning_rate": 0.0004117853560682046, "loss": 2.8117, "theoretical_loss": 3.833593814581876, "tokens_seen": 609550336 }, { "epoch": 0.18, "learning_rate": 0.00041177532597793385, "loss": 2.8192, "theoretical_loss": 3.8335509280895854, "tokens_seen": 609615872 }, { "epoch": 0.18, "learning_rate": 0.000411765295887663, "loss": 2.8, "theoretical_loss": 3.833508047498276, "tokens_seen": 609681408 }, { "epoch": 0.18, "learning_rate": 0.0004117552657973922, "loss": 3.1224, "theoretical_loss": 3.833465172806503, "tokens_seen": 609746944 }, { "epoch": 0.18, "learning_rate": 0.00041174523570712134, "loss": 2.7969, "theoretical_loss": 3.833422304012819, "tokens_seen": 609812480 }, { "epoch": 0.18, "learning_rate": 0.0004117352056168506, "loss": 2.7491, "theoretical_loss": 3.833379441115781, "tokens_seen": 609878016 }, { "epoch": 0.18, "learning_rate": 0.00041172517552657975, "loss": 2.943, "theoretical_loss": 3.833336584113943, "tokens_seen": 609943552 }, { "epoch": 0.18, "learning_rate": 0.00041171514543630894, "loss": 2.9704, "theoretical_loss": 3.8332937330058616, "tokens_seen": 610009088 }, { "epoch": 0.18, "learning_rate": 0.0004117051153460381, "loss": 2.7496, "theoretical_loss": 3.8332508877900935, "tokens_seen": 610074624 }, { "epoch": 0.18, "learning_rate": 0.0004116950852557673, "loss": 2.9887, "theoretical_loss": 3.8332080484651945, "tokens_seen": 610140160 }, { "epoch": 0.18, "learning_rate": 0.0004116850551654965, "loss": 2.7619, "theoretical_loss": 3.8331652150297226, "tokens_seen": 610205696 }, { "epoch": 0.18, "learning_rate": 0.0004116750250752257, "loss": 2.9078, "theoretical_loss": 3.833122387482236, "tokens_seen": 610271232 }, { "epoch": 0.18, "learning_rate": 0.00041166499498495484, "loss": 2.7804, "theoretical_loss": 3.8330795658212926, "tokens_seen": 610336768 }, { "epoch": 0.18, "learning_rate": 0.0004116549648946841, "loss": 2.8807, "theoretical_loss": 3.833036750045452, "tokens_seen": 610402304 }, { "epoch": 0.18, "learning_rate": 0.0004116449348044132, "loss": 2.8797, "theoretical_loss": 3.832993940153273, "tokens_seen": 610467840 }, { "epoch": 0.19, "learning_rate": 0.00041163490471414244, "loss": 2.7678, "theoretical_loss": 3.8329511361433166, "tokens_seen": 610533376 }, { "epoch": 0.19, "learning_rate": 0.0004116248746238716, "loss": 2.8182, "theoretical_loss": 3.832908338014142, "tokens_seen": 610598912 }, { "epoch": 0.19, "learning_rate": 0.0004116148445336008, "loss": 2.6335, "theoretical_loss": 3.832865545764311, "tokens_seen": 610664448 }, { "epoch": 0.19, "learning_rate": 0.00041160481444333, "loss": 2.7076, "theoretical_loss": 3.8328227593923856, "tokens_seen": 610729984 }, { "epoch": 0.19, "learning_rate": 0.0004115947843530592, "loss": 2.9658, "theoretical_loss": 3.832779978896926, "tokens_seen": 610795520 }, { "epoch": 0.19, "learning_rate": 0.00041158475426278834, "loss": 2.7874, "theoretical_loss": 3.832737204276497, "tokens_seen": 610861056 }, { "epoch": 0.19, "learning_rate": 0.0004115747241725176, "loss": 2.7251, "theoretical_loss": 3.8326944355296595, "tokens_seen": 610926592 }, { "epoch": 0.19, "learning_rate": 0.0004115646940822467, "loss": 2.8695, "theoretical_loss": 3.832651672654978, "tokens_seen": 610992128 }, { "epoch": 0.19, "learning_rate": 0.00041155466399197594, "loss": 2.7779, "theoretical_loss": 3.832608915651017, "tokens_seen": 611057664 }, { "epoch": 0.19, "objective/train/docs_used": 356348, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.85497784614563, "objective/train/theoretical_loss": 3.83256616451634, "objective/train/tokens_used": 631583200, "theoretical_loss": 3.83256616451634, "tokens_seen": 611123200 }, { "epoch": 0.19, "learning_rate": 0.0004115446339017051, "loss": 2.8653, "theoretical_loss": 3.83256616451634, "tokens_seen": 611123200 }, { "epoch": 0.19, "learning_rate": 0.0004115346038114343, "loss": 2.6779, "theoretical_loss": 3.8325234192495126, "tokens_seen": 611188736 }, { "epoch": 0.19, "learning_rate": 0.0004115245737211635, "loss": 2.9147, "theoretical_loss": 3.832480679849101, "tokens_seen": 611254272 }, { "epoch": 0.19, "learning_rate": 0.00041151454363089267, "loss": 2.9427, "theoretical_loss": 3.8324379463136697, "tokens_seen": 611319808 }, { "epoch": 0.19, "learning_rate": 0.00041150451354062185, "loss": 2.9044, "theoretical_loss": 3.8323952186417864, "tokens_seen": 611385344 }, { "epoch": 0.19, "learning_rate": 0.0004114944834503511, "loss": 2.9848, "theoretical_loss": 3.832352496832018, "tokens_seen": 611450880 }, { "epoch": 0.19, "learning_rate": 0.0004114844533600802, "loss": 2.8493, "theoretical_loss": 3.832309780882932, "tokens_seen": 611516416 }, { "epoch": 0.19, "learning_rate": 0.00041147442326980944, "loss": 2.6212, "theoretical_loss": 3.8322670707930957, "tokens_seen": 611581952 }, { "epoch": 0.19, "learning_rate": 0.00041146439317953857, "loss": 2.9301, "theoretical_loss": 3.832224366561079, "tokens_seen": 611647488 }, { "epoch": 0.19, "learning_rate": 0.0004114543630892678, "loss": 2.6062, "theoretical_loss": 3.83218166818545, "tokens_seen": 611713024 }, { "epoch": 0.19, "learning_rate": 0.000411444332998997, "loss": 2.7474, "theoretical_loss": 3.8321389756647783, "tokens_seen": 611778560 }, { "epoch": 0.19, "learning_rate": 0.00041143430290872617, "loss": 2.9489, "theoretical_loss": 3.8320962889976347, "tokens_seen": 611844096 }, { "epoch": 0.19, "learning_rate": 0.00041142427281845535, "loss": 2.9554, "theoretical_loss": 3.832053608182589, "tokens_seen": 611909632 }, { "epoch": 0.19, "learning_rate": 0.0004114142427281846, "loss": 2.6481, "theoretical_loss": 3.8320109332182124, "tokens_seen": 611975168 }, { "epoch": 0.19, "learning_rate": 0.00041140421263791377, "loss": 2.9068, "theoretical_loss": 3.831968264103077, "tokens_seen": 612040704 }, { "epoch": 0.19, "learning_rate": 0.00041139418254764295, "loss": 2.779, "theoretical_loss": 3.831925600835754, "tokens_seen": 612106240 }, { "epoch": 0.19, "learning_rate": 0.00041138415245737213, "loss": 2.7237, "theoretical_loss": 3.831882943414817, "tokens_seen": 612171776 }, { "epoch": 0.19, "learning_rate": 0.0004113741223671013, "loss": 3.0293, "theoretical_loss": 3.8318402918388386, "tokens_seen": 612237312 }, { "epoch": 0.19, "learning_rate": 0.00041136409227683055, "loss": 2.7863, "theoretical_loss": 3.8317976461063923, "tokens_seen": 612302848 }, { "epoch": 0.19, "learning_rate": 0.00041135406218655967, "loss": 2.7599, "theoretical_loss": 3.831755006216052, "tokens_seen": 612368384 }, { "epoch": 0.19, "learning_rate": 0.0004113440320962889, "loss": 2.9175, "theoretical_loss": 3.8317123721663924, "tokens_seen": 612433920 }, { "epoch": 0.19, "learning_rate": 0.00041133400200601803, "loss": 2.4928, "theoretical_loss": 3.831669743955989, "tokens_seen": 612499456 }, { "epoch": 0.19, "learning_rate": 0.00041132397191574727, "loss": 2.8791, "theoretical_loss": 3.8316271215834172, "tokens_seen": 612564992 }, { "epoch": 0.19, "learning_rate": 0.00041131394182547645, "loss": 2.7784, "theoretical_loss": 3.8315845050472532, "tokens_seen": 612630528 }, { "epoch": 0.19, "learning_rate": 0.00041130391173520563, "loss": 2.7013, "theoretical_loss": 3.831541894346073, "tokens_seen": 612696064 }, { "epoch": 0.19, "objective/train/docs_used": 356825, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0833096504211426, "objective/train/theoretical_loss": 3.8314992894784536, "objective/train/tokens_used": 633221600, "theoretical_loss": 3.8314992894784536, "tokens_seen": 612761600 }, { "epoch": 0.19, "learning_rate": 0.0004112938816449348, "loss": 2.7547, "theoretical_loss": 3.8314992894784536, "tokens_seen": 612761600 }, { "epoch": 0.19, "learning_rate": 0.00041128385155466405, "loss": 3.0618, "theoretical_loss": 3.831456690442974, "tokens_seen": 612827136 }, { "epoch": 0.19, "learning_rate": 0.0004112738214643932, "loss": 3.0582, "theoretical_loss": 3.8314140972382105, "tokens_seen": 612892672 }, { "epoch": 0.19, "learning_rate": 0.0004112637913741224, "loss": 2.9548, "theoretical_loss": 3.831371509862743, "tokens_seen": 612958208 }, { "epoch": 0.19, "learning_rate": 0.00041125376128385154, "loss": 3.0347, "theoretical_loss": 3.8313289283151497, "tokens_seen": 613023744 }, { "epoch": 0.19, "learning_rate": 0.0004112437311935808, "loss": 3.1745, "theoretical_loss": 3.831286352594011, "tokens_seen": 613089280 }, { "epoch": 0.19, "learning_rate": 0.00041123370110330995, "loss": 2.7535, "theoretical_loss": 3.831243782697906, "tokens_seen": 613154816 }, { "epoch": 0.19, "learning_rate": 0.00041122367101303914, "loss": 2.8579, "theoretical_loss": 3.8312012186254156, "tokens_seen": 613220352 }, { "epoch": 0.19, "learning_rate": 0.0004112136409227683, "loss": 2.7409, "theoretical_loss": 3.831158660375122, "tokens_seen": 613285888 }, { "epoch": 0.19, "learning_rate": 0.0004112036108324975, "loss": 2.9041, "theoretical_loss": 3.831116107945605, "tokens_seen": 613351424 }, { "epoch": 0.19, "learning_rate": 0.0004111935807422267, "loss": 2.8949, "theoretical_loss": 3.8310735613354483, "tokens_seen": 613416960 }, { "epoch": 0.19, "learning_rate": 0.0004111835506519559, "loss": 2.9461, "theoretical_loss": 3.831031020543233, "tokens_seen": 613482496 }, { "epoch": 0.19, "learning_rate": 0.00041117352056168504, "loss": 2.8993, "theoretical_loss": 3.830988485567543, "tokens_seen": 613548032 }, { "epoch": 0.19, "learning_rate": 0.0004111634904714143, "loss": 2.7299, "theoretical_loss": 3.8309459564069615, "tokens_seen": 613613568 }, { "epoch": 0.19, "learning_rate": 0.0004111534603811434, "loss": 2.9761, "theoretical_loss": 3.830903433060073, "tokens_seen": 613679104 }, { "epoch": 0.19, "learning_rate": 0.00041114343029087264, "loss": 3.0147, "theoretical_loss": 3.8308609155254616, "tokens_seen": 613744640 }, { "epoch": 0.19, "learning_rate": 0.0004111334002006018, "loss": 2.8001, "theoretical_loss": 3.830818403801713, "tokens_seen": 613810176 }, { "epoch": 0.19, "learning_rate": 0.000411123370110331, "loss": 3.06, "theoretical_loss": 3.8307758978874116, "tokens_seen": 613875712 }, { "epoch": 0.19, "learning_rate": 0.0004111133400200602, "loss": 2.9081, "theoretical_loss": 3.830733397781144, "tokens_seen": 613941248 }, { "epoch": 0.19, "learning_rate": 0.0004111033099297894, "loss": 2.8054, "theoretical_loss": 3.830690903481497, "tokens_seen": 614006784 }, { "epoch": 0.19, "learning_rate": 0.00041109327983951854, "loss": 2.8168, "theoretical_loss": 3.830648414987057, "tokens_seen": 614072320 }, { "epoch": 0.19, "learning_rate": 0.0004110832497492478, "loss": 3.0625, "theoretical_loss": 3.8306059322964128, "tokens_seen": 614137856 }, { "epoch": 0.19, "learning_rate": 0.0004110732196589769, "loss": 2.8266, "theoretical_loss": 3.830563455408151, "tokens_seen": 614203392 }, { "epoch": 0.19, "learning_rate": 0.00041106318956870614, "loss": 2.9967, "theoretical_loss": 3.8305209843208603, "tokens_seen": 614268928 }, { "epoch": 0.19, "learning_rate": 0.0004110531594784353, "loss": 2.8098, "theoretical_loss": 3.8304785190331305, "tokens_seen": 614334464 }, { "epoch": 0.19, "objective/train/docs_used": 357491, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.030144453048706, "objective/train/theoretical_loss": 3.830436059543551, "objective/train/tokens_used": 634860000, "theoretical_loss": 3.830436059543551, "tokens_seen": 614400000 }, { "epoch": 0.19, "learning_rate": 0.0004110431293881645, "loss": 2.9081, "theoretical_loss": 3.830436059543551, "tokens_seen": 614400000 }, { "epoch": 0.19, "learning_rate": 0.0004110330992978937, "loss": 2.8645, "theoretical_loss": 3.830393605850711, "tokens_seen": 614465536 }, { "epoch": 0.19, "learning_rate": 0.00041102306920762287, "loss": 2.6907, "theoretical_loss": 3.8303511579532015, "tokens_seen": 614531072 }, { "epoch": 0.19, "learning_rate": 0.00041101303911735205, "loss": 2.8431, "theoretical_loss": 3.8303087158496134, "tokens_seen": 614596608 }, { "epoch": 0.19, "learning_rate": 0.0004110030090270813, "loss": 2.9891, "theoretical_loss": 3.8302662795385385, "tokens_seen": 614662144 }, { "epoch": 0.19, "learning_rate": 0.0004109929789368104, "loss": 2.9626, "theoretical_loss": 3.8302238490185676, "tokens_seen": 614727680 }, { "epoch": 0.19, "learning_rate": 0.00041098294884653965, "loss": 2.6998, "theoretical_loss": 3.830181424288295, "tokens_seen": 614793216 }, { "epoch": 0.19, "learning_rate": 0.00041097291875626877, "loss": 2.7542, "theoretical_loss": 3.830139005346312, "tokens_seen": 614858752 }, { "epoch": 0.19, "learning_rate": 0.000410962888665998, "loss": 2.7335, "theoretical_loss": 3.830096592191212, "tokens_seen": 614924288 }, { "epoch": 0.19, "learning_rate": 0.0004109528585757272, "loss": 2.7658, "theoretical_loss": 3.8300541848215905, "tokens_seen": 614989824 }, { "epoch": 0.19, "learning_rate": 0.00041094282848545637, "loss": 2.7634, "theoretical_loss": 3.8300117832360403, "tokens_seen": 615055360 }, { "epoch": 0.19, "learning_rate": 0.00041093279839518555, "loss": 2.8337, "theoretical_loss": 3.8299693874331573, "tokens_seen": 615120896 }, { "epoch": 0.19, "learning_rate": 0.0004109227683049148, "loss": 2.8861, "theoretical_loss": 3.829926997411537, "tokens_seen": 615186432 }, { "epoch": 0.19, "learning_rate": 0.0004109127382146439, "loss": 2.9835, "theoretical_loss": 3.8298846131697744, "tokens_seen": 615251968 }, { "epoch": 0.19, "learning_rate": 0.00041090270812437315, "loss": 2.8007, "theoretical_loss": 3.829842234706466, "tokens_seen": 615317504 }, { "epoch": 0.19, "learning_rate": 0.0004108926780341023, "loss": 2.9777, "theoretical_loss": 3.8297998620202094, "tokens_seen": 615383040 }, { "epoch": 0.19, "learning_rate": 0.0004108826479438315, "loss": 2.7049, "theoretical_loss": 3.829757495109602, "tokens_seen": 615448576 }, { "epoch": 0.19, "learning_rate": 0.0004108726178535607, "loss": 2.7635, "theoretical_loss": 3.8297151339732407, "tokens_seen": 615514112 }, { "epoch": 0.19, "learning_rate": 0.00041086258776328987, "loss": 2.9916, "theoretical_loss": 3.8296727786097247, "tokens_seen": 615579648 }, { "epoch": 0.19, "learning_rate": 0.00041085255767301905, "loss": 2.9294, "theoretical_loss": 3.8296304290176524, "tokens_seen": 615645184 }, { "epoch": 0.19, "learning_rate": 0.00041084252758274823, "loss": 2.9406, "theoretical_loss": 3.829588085195623, "tokens_seen": 615710720 }, { "epoch": 0.19, "learning_rate": 0.0004108324974924774, "loss": 3.1951, "theoretical_loss": 3.8295457471422365, "tokens_seen": 615776256 }, { "epoch": 0.19, "learning_rate": 0.00041082246740220665, "loss": 2.8901, "theoretical_loss": 3.8295034148560934, "tokens_seen": 615841792 }, { "epoch": 0.19, "learning_rate": 0.0004108124373119358, "loss": 2.7115, "theoretical_loss": 3.8294610883357945, "tokens_seen": 615907328 }, { "epoch": 0.19, "learning_rate": 0.000410802407221665, "loss": 2.8682, "theoretical_loss": 3.8294187675799405, "tokens_seen": 615972864 }, { "epoch": 0.19, "objective/train/docs_used": 358741, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9842801094055176, "objective/train/theoretical_loss": 3.829376452587134, "objective/train/tokens_used": 636498400, "theoretical_loss": 3.829376452587134, "tokens_seen": 616038400 }, { "epoch": 0.19, "learning_rate": 0.0004107923771313942, "loss": 2.756, "theoretical_loss": 3.829376452587134, "tokens_seen": 616038400 }, { "epoch": 0.19, "learning_rate": 0.0004107823470411234, "loss": 2.7453, "theoretical_loss": 3.8293341433559767, "tokens_seen": 616103936 }, { "epoch": 0.19, "learning_rate": 0.00041077231695085256, "loss": 2.7221, "theoretical_loss": 3.8292918398850713, "tokens_seen": 616169472 }, { "epoch": 0.19, "learning_rate": 0.00041076228686058174, "loss": 3.0387, "theoretical_loss": 3.829249542173021, "tokens_seen": 616235008 }, { "epoch": 0.19, "learning_rate": 0.0004107522567703109, "loss": 2.8791, "theoretical_loss": 3.82920725021843, "tokens_seen": 616300544 }, { "epoch": 0.19, "learning_rate": 0.00041074222668004015, "loss": 2.8977, "theoretical_loss": 3.8291649640199017, "tokens_seen": 616366080 }, { "epoch": 0.19, "learning_rate": 0.0004107321965897693, "loss": 2.974, "theoretical_loss": 3.829122683576042, "tokens_seen": 616431616 }, { "epoch": 0.19, "learning_rate": 0.0004107221664994985, "loss": 3.0914, "theoretical_loss": 3.8290804088854546, "tokens_seen": 616497152 }, { "epoch": 0.19, "learning_rate": 0.00041071213640922764, "loss": 2.7757, "theoretical_loss": 3.8290381399467464, "tokens_seen": 616562688 }, { "epoch": 0.19, "learning_rate": 0.0004107021063189569, "loss": 2.9125, "theoretical_loss": 3.8289958767585226, "tokens_seen": 616628224 }, { "epoch": 0.19, "learning_rate": 0.00041069207622868606, "loss": 2.76, "theoretical_loss": 3.8289536193193907, "tokens_seen": 616693760 }, { "epoch": 0.19, "learning_rate": 0.00041068204613841524, "loss": 2.7955, "theoretical_loss": 3.828911367627957, "tokens_seen": 616759296 }, { "epoch": 0.19, "learning_rate": 0.0004106720160481444, "loss": 2.8873, "theoretical_loss": 3.8288691216828297, "tokens_seen": 616824832 }, { "epoch": 0.19, "learning_rate": 0.0004106619859578736, "loss": 2.6994, "theoretical_loss": 3.828826881482617, "tokens_seen": 616890368 }, { "epoch": 0.19, "learning_rate": 0.00041065195586760284, "loss": 2.8009, "theoretical_loss": 3.8287846470259264, "tokens_seen": 616955904 }, { "epoch": 0.19, "learning_rate": 0.000410641925777332, "loss": 2.7532, "theoretical_loss": 3.828742418311368, "tokens_seen": 617021440 }, { "epoch": 0.19, "learning_rate": 0.0004106318956870612, "loss": 2.768, "theoretical_loss": 3.828700195337551, "tokens_seen": 617086976 }, { "epoch": 0.19, "learning_rate": 0.0004106218655967904, "loss": 2.9357, "theoretical_loss": 3.8286579781030863, "tokens_seen": 617152512 }, { "epoch": 0.19, "learning_rate": 0.0004106118355065196, "loss": 2.9161, "theoretical_loss": 3.8286157666065828, "tokens_seen": 617218048 }, { "epoch": 0.19, "learning_rate": 0.00041060180541624874, "loss": 2.932, "theoretical_loss": 3.828573560846652, "tokens_seen": 617283584 }, { "epoch": 0.19, "learning_rate": 0.000410591775325978, "loss": 2.9844, "theoretical_loss": 3.8285313608219056, "tokens_seen": 617349120 }, { "epoch": 0.19, "learning_rate": 0.0004105817452357071, "loss": 2.7526, "theoretical_loss": 3.8284891665309564, "tokens_seen": 617414656 }, { "epoch": 0.19, "learning_rate": 0.00041057171514543634, "loss": 2.7736, "theoretical_loss": 3.8284469779724155, "tokens_seen": 617480192 }, { "epoch": 0.19, "learning_rate": 0.0004105616850551655, "loss": 2.8821, "theoretical_loss": 3.8284047951448965, "tokens_seen": 617545728 }, { "epoch": 0.19, "learning_rate": 0.0004105516549648947, "loss": 2.8562, "theoretical_loss": 3.828362618047013, "tokens_seen": 617611264 }, { "epoch": 0.19, "objective/train/docs_used": 359404, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.935211181640625, "objective/train/theoretical_loss": 3.828320446677378, "objective/train/tokens_used": 638136800, "theoretical_loss": 3.828320446677378, "tokens_seen": 617676800 }, { "epoch": 0.19, "learning_rate": 0.0004105416248746239, "loss": 2.5811, "theoretical_loss": 3.828320446677378, "tokens_seen": 617676800 }, { "epoch": 0.19, "learning_rate": 0.00041053159478435307, "loss": 2.8761, "theoretical_loss": 3.8282782810346063, "tokens_seen": 617742336 }, { "epoch": 0.19, "learning_rate": 0.00041052156469408225, "loss": 3.19, "theoretical_loss": 3.8282361211173135, "tokens_seen": 617807872 }, { "epoch": 0.19, "learning_rate": 0.0004105115346038115, "loss": 2.6977, "theoretical_loss": 3.8281939669241143, "tokens_seen": 617873408 }, { "epoch": 0.19, "learning_rate": 0.0004105015045135406, "loss": 3.0995, "theoretical_loss": 3.8281518184536245, "tokens_seen": 617938944 }, { "epoch": 0.19, "learning_rate": 0.00041049147442326985, "loss": 2.9678, "theoretical_loss": 3.82810967570446, "tokens_seen": 618004480 }, { "epoch": 0.19, "learning_rate": 0.00041048144433299897, "loss": 2.8602, "theoretical_loss": 3.8280675386752385, "tokens_seen": 618070016 }, { "epoch": 0.19, "learning_rate": 0.0004104714142427282, "loss": 2.9506, "theoretical_loss": 3.828025407364577, "tokens_seen": 618135552 }, { "epoch": 0.19, "learning_rate": 0.0004104613841524574, "loss": 2.7122, "theoretical_loss": 3.827983281771093, "tokens_seen": 618201088 }, { "epoch": 0.19, "learning_rate": 0.00041045135406218657, "loss": 2.6051, "theoretical_loss": 3.8279411618934045, "tokens_seen": 618266624 }, { "epoch": 0.19, "learning_rate": 0.00041044132397191575, "loss": 3.041, "theoretical_loss": 3.8278990477301305, "tokens_seen": 618332160 }, { "epoch": 0.19, "learning_rate": 0.000410431293881645, "loss": 2.7328, "theoretical_loss": 3.82785693927989, "tokens_seen": 618397696 }, { "epoch": 0.19, "learning_rate": 0.0004104212637913741, "loss": 2.7715, "theoretical_loss": 3.827814836541303, "tokens_seen": 618463232 }, { "epoch": 0.19, "learning_rate": 0.00041041123370110335, "loss": 2.9858, "theoretical_loss": 3.82777273951299, "tokens_seen": 618528768 }, { "epoch": 0.19, "learning_rate": 0.0004104012036108325, "loss": 2.8936, "theoretical_loss": 3.8277306481935707, "tokens_seen": 618594304 }, { "epoch": 0.19, "learning_rate": 0.0004103911735205617, "loss": 2.9234, "theoretical_loss": 3.8276885625816663, "tokens_seen": 618659840 }, { "epoch": 0.19, "learning_rate": 0.0004103811434302909, "loss": 2.7475, "theoretical_loss": 3.827646482675899, "tokens_seen": 618725376 }, { "epoch": 0.19, "learning_rate": 0.00041037111334002007, "loss": 3.0746, "theoretical_loss": 3.8276044084748904, "tokens_seen": 618790912 }, { "epoch": 0.19, "learning_rate": 0.00041036108324974925, "loss": 2.7544, "theoretical_loss": 3.8275623399772636, "tokens_seen": 618856448 }, { "epoch": 0.19, "learning_rate": 0.00041035105315947844, "loss": 2.4142, "theoretical_loss": 3.827520277181641, "tokens_seen": 618921984 }, { "epoch": 0.19, "learning_rate": 0.0004103410230692076, "loss": 2.9755, "theoretical_loss": 3.8274782200866464, "tokens_seen": 618987520 }, { "epoch": 0.19, "learning_rate": 0.00041033099297893685, "loss": 2.8222, "theoretical_loss": 3.8274361686909035, "tokens_seen": 619053056 }, { "epoch": 0.19, "learning_rate": 0.000410320962888666, "loss": 2.9724, "theoretical_loss": 3.827394122993037, "tokens_seen": 619118592 }, { "epoch": 0.19, "learning_rate": 0.0004103109327983952, "loss": 2.8992, "theoretical_loss": 3.827352082991672, "tokens_seen": 619184128 }, { "epoch": 0.19, "learning_rate": 0.0004103009027081244, "loss": 2.8078, "theoretical_loss": 3.8273100486854332, "tokens_seen": 619249664 }, { "epoch": 0.19, "objective/train/docs_used": 360794, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5488815307617188, "objective/train/theoretical_loss": 3.827268020072948, "objective/train/tokens_used": 639775200, "theoretical_loss": 3.827268020072948, "tokens_seen": 619315200 }, { "epoch": 0.19, "learning_rate": 0.0004102908726178536, "loss": 2.8808, "theoretical_loss": 3.827268020072948, "tokens_seen": 619315200 }, { "epoch": 0.19, "learning_rate": 0.00041028084252758276, "loss": 2.8708, "theoretical_loss": 3.8272259971528406, "tokens_seen": 619380736 }, { "epoch": 0.19, "learning_rate": 0.00041027081243731194, "loss": 2.7716, "theoretical_loss": 3.8271839799237397, "tokens_seen": 619446272 }, { "epoch": 0.19, "learning_rate": 0.0004102607823470411, "loss": 2.7464, "theoretical_loss": 3.8271419683842716, "tokens_seen": 619511808 }, { "epoch": 0.19, "learning_rate": 0.00041025075225677035, "loss": 2.6849, "theoretical_loss": 3.8270999625330644, "tokens_seen": 619577344 }, { "epoch": 0.19, "learning_rate": 0.0004102407221664995, "loss": 2.963, "theoretical_loss": 3.8270579623687464, "tokens_seen": 619642880 }, { "epoch": 0.19, "learning_rate": 0.0004102306920762287, "loss": 2.789, "theoretical_loss": 3.8270159678899462, "tokens_seen": 619708416 }, { "epoch": 0.19, "learning_rate": 0.00041022066198595784, "loss": 2.7791, "theoretical_loss": 3.826973979095293, "tokens_seen": 619773952 }, { "epoch": 0.19, "learning_rate": 0.0004102106318956871, "loss": 2.9164, "theoretical_loss": 3.8269319959834167, "tokens_seen": 619839488 }, { "epoch": 0.19, "learning_rate": 0.00041020060180541626, "loss": 2.74, "theoretical_loss": 3.8268900185529473, "tokens_seen": 619905024 }, { "epoch": 0.19, "learning_rate": 0.00041019057171514544, "loss": 2.7933, "theoretical_loss": 3.8268480468025157, "tokens_seen": 619970560 }, { "epoch": 0.19, "learning_rate": 0.0004101805416248746, "loss": 2.7509, "theoretical_loss": 3.826806080730753, "tokens_seen": 620036096 }, { "epoch": 0.19, "learning_rate": 0.0004101705115346038, "loss": 2.683, "theoretical_loss": 3.82676412033629, "tokens_seen": 620101632 }, { "epoch": 0.19, "learning_rate": 0.000410160481444333, "loss": 2.4988, "theoretical_loss": 3.8267221656177606, "tokens_seen": 620167168 }, { "epoch": 0.19, "learning_rate": 0.0004101504513540622, "loss": 2.9297, "theoretical_loss": 3.826680216573795, "tokens_seen": 620232704 }, { "epoch": 0.19, "learning_rate": 0.00041014042126379135, "loss": 2.8455, "theoretical_loss": 3.8266382732030277, "tokens_seen": 620298240 }, { "epoch": 0.19, "learning_rate": 0.0004101303911735206, "loss": 2.5712, "theoretical_loss": 3.8265963355040915, "tokens_seen": 620363776 }, { "epoch": 0.19, "learning_rate": 0.00041012036108324976, "loss": 2.7332, "theoretical_loss": 3.8265544034756216, "tokens_seen": 620429312 }, { "epoch": 0.19, "learning_rate": 0.00041011033099297894, "loss": 2.8088, "theoretical_loss": 3.8265124771162506, "tokens_seen": 620494848 }, { "epoch": 0.19, "learning_rate": 0.0004101003009027081, "loss": 2.8373, "theoretical_loss": 3.8264705564246153, "tokens_seen": 620560384 }, { "epoch": 0.19, "learning_rate": 0.0004100902708124373, "loss": 2.8893, "theoretical_loss": 3.8264286413993496, "tokens_seen": 620625920 }, { "epoch": 0.19, "learning_rate": 0.0004100802407221665, "loss": 2.7777, "theoretical_loss": 3.82638673203909, "tokens_seen": 620691456 }, { "epoch": 0.19, "learning_rate": 0.0004100702106318957, "loss": 2.8751, "theoretical_loss": 3.8263448283424726, "tokens_seen": 620756992 }, { "epoch": 0.19, "learning_rate": 0.00041006018054162485, "loss": 2.9087, "theoretical_loss": 3.8263029303081346, "tokens_seen": 620822528 }, { "epoch": 0.19, "learning_rate": 0.0004100501504513541, "loss": 2.9091, "theoretical_loss": 3.826261037934713, "tokens_seen": 620888064 }, { "epoch": 0.19, "objective/train/docs_used": 361341, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.479189395904541, "objective/train/theoretical_loss": 3.8262191512208448, "objective/train/tokens_used": 641413600, "theoretical_loss": 3.8262191512208448, "tokens_seen": 620953600 }, { "epoch": 0.19, "learning_rate": 0.0004100401203610832, "loss": 2.7642, "theoretical_loss": 3.8262191512208448, "tokens_seen": 620953600 }, { "epoch": 0.19, "learning_rate": 0.00041003009027081245, "loss": 3.0473, "theoretical_loss": 3.8261772701651697, "tokens_seen": 621019136 }, { "epoch": 0.19, "learning_rate": 0.00041002006018054163, "loss": 2.8635, "theoretical_loss": 3.8261353947663252, "tokens_seen": 621084672 }, { "epoch": 0.19, "learning_rate": 0.0004100100300902708, "loss": 2.5395, "theoretical_loss": 3.8260935250229515, "tokens_seen": 621150208 }, { "epoch": 0.19, "learning_rate": 0.00041, "loss": 2.8946, "theoretical_loss": 3.826051660933687, "tokens_seen": 621215744 }, { "epoch": 0.19, "learning_rate": 0.00040998996990972917, "loss": 2.883, "theoretical_loss": 3.8260098024971727, "tokens_seen": 621281280 }, { "epoch": 0.19, "learning_rate": 0.00040997993981945835, "loss": 2.9444, "theoretical_loss": 3.825967949712049, "tokens_seen": 621346816 }, { "epoch": 0.19, "learning_rate": 0.0004099699097291876, "loss": 2.9486, "theoretical_loss": 3.8259261025769566, "tokens_seen": 621412352 }, { "epoch": 0.19, "learning_rate": 0.0004099598796389167, "loss": 3.0868, "theoretical_loss": 3.8258842610905375, "tokens_seen": 621477888 }, { "epoch": 0.19, "learning_rate": 0.00040994984954864595, "loss": 2.8341, "theoretical_loss": 3.825842425251434, "tokens_seen": 621543424 }, { "epoch": 0.19, "learning_rate": 0.00040993981945837513, "loss": 2.9232, "theoretical_loss": 3.8258005950582867, "tokens_seen": 621608960 }, { "epoch": 0.19, "learning_rate": 0.0004099297893681043, "loss": 3.089, "theoretical_loss": 3.8257587705097404, "tokens_seen": 621674496 }, { "epoch": 0.19, "learning_rate": 0.0004099197592778335, "loss": 2.8746, "theoretical_loss": 3.825716951604438, "tokens_seen": 621740032 }, { "epoch": 0.19, "learning_rate": 0.0004099097291875627, "loss": 2.9131, "theoretical_loss": 3.8256751383410235, "tokens_seen": 621805568 }, { "epoch": 0.19, "learning_rate": 0.0004098996990972919, "loss": 2.8811, "theoretical_loss": 3.82563333071814, "tokens_seen": 621871104 }, { "epoch": 0.19, "learning_rate": 0.0004098896690070211, "loss": 2.8428, "theoretical_loss": 3.825591528734434, "tokens_seen": 621936640 }, { "epoch": 0.19, "learning_rate": 0.0004098796389167503, "loss": 2.8625, "theoretical_loss": 3.82554973238855, "tokens_seen": 622002176 }, { "epoch": 0.19, "learning_rate": 0.00040986960882647945, "loss": 2.6699, "theoretical_loss": 3.8255079416791338, "tokens_seen": 622067712 }, { "epoch": 0.19, "learning_rate": 0.00040985957873620864, "loss": 2.7263, "theoretical_loss": 3.825466156604831, "tokens_seen": 622133248 }, { "epoch": 0.19, "learning_rate": 0.0004098495486459378, "loss": 2.9442, "theoretical_loss": 3.8254243771642895, "tokens_seen": 622198784 }, { "epoch": 0.19, "learning_rate": 0.00040983951855566705, "loss": 2.7449, "theoretical_loss": 3.8253826033561555, "tokens_seen": 622264320 }, { "epoch": 0.19, "learning_rate": 0.0004098294884653962, "loss": 2.8408, "theoretical_loss": 3.8253408351790767, "tokens_seen": 622329856 }, { "epoch": 0.19, "learning_rate": 0.0004098194583751254, "loss": 2.6127, "theoretical_loss": 3.8252990726317018, "tokens_seen": 622395392 }, { "epoch": 0.19, "learning_rate": 0.0004098094282848546, "loss": 3.1112, "theoretical_loss": 3.825257315712678, "tokens_seen": 622460928 }, { "epoch": 0.19, "learning_rate": 0.0004097993981945838, "loss": 2.7346, "theoretical_loss": 3.825215564420656, "tokens_seen": 622526464 }, { "epoch": 0.19, "objective/train/docs_used": 362042, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1056301593780518, "objective/train/theoretical_loss": 3.8251738187542843, "objective/train/tokens_used": 643052000, "theoretical_loss": 3.8251738187542843, "tokens_seen": 622592000 }, { "epoch": 0.19, "learning_rate": 0.00040978936810431296, "loss": 2.9863, "theoretical_loss": 3.8251738187542843, "tokens_seen": 622592000 }, { "epoch": 0.19, "learning_rate": 0.00040977933801404214, "loss": 2.7314, "theoretical_loss": 3.8251320787122127, "tokens_seen": 622657536 }, { "epoch": 0.19, "learning_rate": 0.0004097693079237713, "loss": 2.7175, "theoretical_loss": 3.8250903442930917, "tokens_seen": 622723072 }, { "epoch": 0.19, "learning_rate": 0.00040975927783350055, "loss": 2.7428, "theoretical_loss": 3.8250486154955725, "tokens_seen": 622788608 }, { "epoch": 0.19, "learning_rate": 0.0004097492477432297, "loss": 2.9722, "theoretical_loss": 3.8250068923183065, "tokens_seen": 622854144 }, { "epoch": 0.19, "learning_rate": 0.0004097392176529589, "loss": 2.815, "theoretical_loss": 3.824965174759945, "tokens_seen": 622919680 }, { "epoch": 0.19, "learning_rate": 0.00040972918756268804, "loss": 2.7863, "theoretical_loss": 3.8249234628191404, "tokens_seen": 622985216 }, { "epoch": 0.19, "learning_rate": 0.0004097191574724173, "loss": 2.8445, "theoretical_loss": 3.8248817564945465, "tokens_seen": 623050752 }, { "epoch": 0.19, "learning_rate": 0.00040970912738214646, "loss": 2.9068, "theoretical_loss": 3.8248400557848146, "tokens_seen": 623116288 }, { "epoch": 0.19, "learning_rate": 0.00040969909729187564, "loss": 2.7792, "theoretical_loss": 3.8247983606885994, "tokens_seen": 623181824 }, { "epoch": 0.19, "learning_rate": 0.0004096890672016048, "loss": 2.8544, "theoretical_loss": 3.824756671204555, "tokens_seen": 623247360 }, { "epoch": 0.19, "learning_rate": 0.000409679037111334, "loss": 2.6811, "theoretical_loss": 3.824714987331336, "tokens_seen": 623312896 }, { "epoch": 0.19, "learning_rate": 0.0004096690070210632, "loss": 3.0151, "theoretical_loss": 3.824673309067598, "tokens_seen": 623378432 }, { "epoch": 0.19, "learning_rate": 0.0004096589769307924, "loss": 2.8711, "theoretical_loss": 3.8246316364119948, "tokens_seen": 623443968 }, { "epoch": 0.19, "learning_rate": 0.00040964894684052155, "loss": 2.8549, "theoretical_loss": 3.824589969363184, "tokens_seen": 623509504 }, { "epoch": 0.19, "learning_rate": 0.0004096389167502508, "loss": 2.7348, "theoretical_loss": 3.8245483079198213, "tokens_seen": 623575040 }, { "epoch": 0.19, "learning_rate": 0.00040962888665997996, "loss": 2.5999, "theoretical_loss": 3.824506652080564, "tokens_seen": 623640576 }, { "epoch": 0.19, "learning_rate": 0.00040961885656970914, "loss": 2.8808, "theoretical_loss": 3.824465001844069, "tokens_seen": 623706112 }, { "epoch": 0.19, "learning_rate": 0.0004096088264794383, "loss": 3.0079, "theoretical_loss": 3.8244233572089943, "tokens_seen": 623771648 }, { "epoch": 0.19, "learning_rate": 0.0004095987963891675, "loss": 2.985, "theoretical_loss": 3.824381718173999, "tokens_seen": 623837184 }, { "epoch": 0.19, "learning_rate": 0.0004095887662988967, "loss": 3.093, "theoretical_loss": 3.8243400847377407, "tokens_seen": 623902720 }, { "epoch": 0.19, "learning_rate": 0.0004095787362086259, "loss": 2.8866, "theoretical_loss": 3.824298456898879, "tokens_seen": 623968256 }, { "epoch": 0.19, "learning_rate": 0.00040956870611835505, "loss": 2.8234, "theoretical_loss": 3.8242568346560737, "tokens_seen": 624033792 }, { "epoch": 0.19, "learning_rate": 0.0004095586760280843, "loss": 2.8825, "theoretical_loss": 3.8242152180079847, "tokens_seen": 624099328 }, { "epoch": 0.19, "learning_rate": 0.0004095486459378134, "loss": 2.6858, "theoretical_loss": 3.8241736069532726, "tokens_seen": 624164864 }, { "epoch": 0.19, "objective/train/docs_used": 363016, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.2174317836761475, "objective/train/theoretical_loss": 3.8241320014905993, "objective/train/tokens_used": 644690400, "theoretical_loss": 3.8241320014905993, "tokens_seen": 624230400 }, { "epoch": 0.19, "learning_rate": 0.00040953861584754265, "loss": 2.76, "theoretical_loss": 3.8241320014905993, "tokens_seen": 624230400 }, { "epoch": 0.19, "learning_rate": 0.00040952858575727183, "loss": 2.7017, "theoretical_loss": 3.824090401618625, "tokens_seen": 624295936 }, { "epoch": 0.19, "learning_rate": 0.000409518555667001, "loss": 3.1082, "theoretical_loss": 3.8240488073360126, "tokens_seen": 624361472 }, { "epoch": 0.19, "learning_rate": 0.0004095085255767302, "loss": 2.6126, "theoretical_loss": 3.8240072186414245, "tokens_seen": 624427008 }, { "epoch": 0.19, "learning_rate": 0.00040949849548645937, "loss": 3.0417, "theoretical_loss": 3.8239656355335234, "tokens_seen": 624492544 }, { "epoch": 0.19, "learning_rate": 0.00040948846539618855, "loss": 3.148, "theoretical_loss": 3.8239240580109723, "tokens_seen": 624558080 }, { "epoch": 0.19, "learning_rate": 0.0004094784353059178, "loss": 2.7648, "theoretical_loss": 3.8238824860724354, "tokens_seen": 624623616 }, { "epoch": 0.19, "learning_rate": 0.0004094684052156469, "loss": 2.9155, "theoretical_loss": 3.8238409197165772, "tokens_seen": 624689152 }, { "epoch": 0.19, "learning_rate": 0.00040945837512537615, "loss": 2.7267, "theoretical_loss": 3.823799358942062, "tokens_seen": 624754688 }, { "epoch": 0.19, "learning_rate": 0.00040944834503510533, "loss": 2.978, "theoretical_loss": 3.8237578037475553, "tokens_seen": 624820224 }, { "epoch": 0.19, "learning_rate": 0.0004094383149448345, "loss": 2.7433, "theoretical_loss": 3.823716254131723, "tokens_seen": 624885760 }, { "epoch": 0.19, "learning_rate": 0.0004094282848545637, "loss": 2.8681, "theoretical_loss": 3.823674710093231, "tokens_seen": 624951296 }, { "epoch": 0.19, "learning_rate": 0.0004094182547642929, "loss": 3.0491, "theoretical_loss": 3.8236331716307452, "tokens_seen": 625016832 }, { "epoch": 0.19, "learning_rate": 0.00040940822467402206, "loss": 2.9378, "theoretical_loss": 3.8235916387429336, "tokens_seen": 625082368 }, { "epoch": 0.19, "learning_rate": 0.0004093981945837513, "loss": 2.57, "theoretical_loss": 3.823550111428463, "tokens_seen": 625147904 }, { "epoch": 0.19, "learning_rate": 0.0004093881644934804, "loss": 2.7938, "theoretical_loss": 3.8235085896860017, "tokens_seen": 625213440 }, { "epoch": 0.19, "learning_rate": 0.00040937813440320965, "loss": 3.0367, "theoretical_loss": 3.8234670735142187, "tokens_seen": 625278976 }, { "epoch": 0.19, "learning_rate": 0.0004093681043129388, "loss": 2.8386, "theoretical_loss": 3.8234255629117815, "tokens_seen": 625344512 }, { "epoch": 0.19, "learning_rate": 0.000409358074222668, "loss": 2.6205, "theoretical_loss": 3.8233840578773606, "tokens_seen": 625410048 }, { "epoch": 0.19, "learning_rate": 0.0004093480441323972, "loss": 2.9303, "theoretical_loss": 3.8233425584096254, "tokens_seen": 625475584 }, { "epoch": 0.19, "learning_rate": 0.0004093380140421264, "loss": 2.8171, "theoretical_loss": 3.8233010645072456, "tokens_seen": 625541120 }, { "epoch": 0.19, "learning_rate": 0.00040932798395185556, "loss": 2.9981, "theoretical_loss": 3.8232595761688932, "tokens_seen": 625606656 }, { "epoch": 0.19, "learning_rate": 0.0004093179538615848, "loss": 2.697, "theoretical_loss": 3.8232180933932383, "tokens_seen": 625672192 }, { "epoch": 0.19, "learning_rate": 0.0004093079237713139, "loss": 2.6989, "theoretical_loss": 3.8231766161789524, "tokens_seen": 625737728 }, { "epoch": 0.19, "learning_rate": 0.00040929789368104316, "loss": 2.7926, "theoretical_loss": 3.8231351445247084, "tokens_seen": 625803264 }, { "epoch": 0.19, "objective/train/docs_used": 363672, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9359257221221924, "objective/train/theoretical_loss": 3.8230936784291787, "objective/train/tokens_used": 646328800, "theoretical_loss": 3.8230936784291787, "tokens_seen": 625868800 }, { "epoch": 0.19, "learning_rate": 0.0004092878635907723, "loss": 2.8781, "theoretical_loss": 3.8230936784291787, "tokens_seen": 625868800 }, { "epoch": 0.19, "learning_rate": 0.0004092778335005015, "loss": 2.9496, "theoretical_loss": 3.8230522178910356, "tokens_seen": 625934336 }, { "epoch": 0.19, "learning_rate": 0.0004092678034102307, "loss": 2.9525, "theoretical_loss": 3.823010762908953, "tokens_seen": 625999872 }, { "epoch": 0.19, "learning_rate": 0.0004092577733199599, "loss": 2.7951, "theoretical_loss": 3.8229693134816043, "tokens_seen": 626065408 }, { "epoch": 0.19, "learning_rate": 0.00040924774322968906, "loss": 2.9869, "theoretical_loss": 3.822927869607664, "tokens_seen": 626130944 }, { "epoch": 0.19, "learning_rate": 0.00040923771313941824, "loss": 2.7103, "theoretical_loss": 3.8228864312858084, "tokens_seen": 626196480 }, { "epoch": 0.19, "learning_rate": 0.0004092276830491474, "loss": 2.9631, "theoretical_loss": 3.822844998514711, "tokens_seen": 626262016 }, { "epoch": 0.19, "learning_rate": 0.00040921765295887666, "loss": 3.0385, "theoretical_loss": 3.822803571293048, "tokens_seen": 626327552 }, { "epoch": 0.19, "learning_rate": 0.0004092076228686058, "loss": 2.9119, "theoretical_loss": 3.8227621496194946, "tokens_seen": 626393088 }, { "epoch": 0.19, "learning_rate": 0.000409197592778335, "loss": 2.7128, "theoretical_loss": 3.8227207334927296, "tokens_seen": 626458624 }, { "epoch": 0.19, "learning_rate": 0.00040918756268806415, "loss": 2.9564, "theoretical_loss": 3.822679322911429, "tokens_seen": 626524160 }, { "epoch": 0.19, "learning_rate": 0.0004091775325977934, "loss": 2.8966, "theoretical_loss": 3.8226379178742693, "tokens_seen": 626589696 }, { "epoch": 0.19, "learning_rate": 0.00040916750250752257, "loss": 3.1937, "theoretical_loss": 3.82259651837993, "tokens_seen": 626655232 }, { "epoch": 0.19, "learning_rate": 0.00040915747241725175, "loss": 2.8648, "theoretical_loss": 3.822555124427089, "tokens_seen": 626720768 }, { "epoch": 0.19, "learning_rate": 0.000409147442326981, "loss": 2.7456, "theoretical_loss": 3.8225137360144252, "tokens_seen": 626786304 }, { "epoch": 0.19, "learning_rate": 0.00040913741223671016, "loss": 2.8642, "theoretical_loss": 3.8224723531406175, "tokens_seen": 626851840 }, { "epoch": 0.19, "learning_rate": 0.00040912738214643934, "loss": 3.0406, "theoretical_loss": 3.8224309758043464, "tokens_seen": 626917376 }, { "epoch": 0.19, "learning_rate": 0.0004091173520561685, "loss": 2.8214, "theoretical_loss": 3.8223896040042913, "tokens_seen": 626982912 }, { "epoch": 0.19, "learning_rate": 0.0004091073219658977, "loss": 3.0053, "theoretical_loss": 3.822348237739134, "tokens_seen": 627048448 }, { "epoch": 0.19, "learning_rate": 0.0004090972918756269, "loss": 3.0134, "theoretical_loss": 3.8223068770075552, "tokens_seen": 627113984 }, { "epoch": 0.19, "learning_rate": 0.0004090872617853561, "loss": 2.8663, "theoretical_loss": 3.822265521808236, "tokens_seen": 627179520 }, { "epoch": 0.19, "learning_rate": 0.00040907723169508525, "loss": 3.1141, "theoretical_loss": 3.822224172139859, "tokens_seen": 627245056 }, { "epoch": 0.19, "learning_rate": 0.0004090672016048145, "loss": 2.8602, "theoretical_loss": 3.8221828280011065, "tokens_seen": 627310592 }, { "epoch": 0.19, "learning_rate": 0.0004090571715145436, "loss": 2.9019, "theoretical_loss": 3.8221414893906616, "tokens_seen": 627376128 }, { "epoch": 0.19, "learning_rate": 0.00040904714142427285, "loss": 2.9343, "theoretical_loss": 3.822100156307208, "tokens_seen": 627441664 }, { "epoch": 0.19, "objective/train/docs_used": 364826, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7832794189453125, "objective/train/theoretical_loss": 3.8220588287494284, "objective/train/tokens_used": 647967200, "theoretical_loss": 3.8220588287494284, "tokens_seen": 627507200 }, { "epoch": 0.19, "learning_rate": 0.00040903711133400203, "loss": 2.8941, "theoretical_loss": 3.8220588287494284, "tokens_seen": 627507200 }, { "epoch": 0.19, "learning_rate": 0.0004090270812437312, "loss": 3.0035, "theoretical_loss": 3.8220175067160085, "tokens_seen": 627572736 }, { "epoch": 0.19, "learning_rate": 0.0004090170511534604, "loss": 2.9709, "theoretical_loss": 3.821976190205632, "tokens_seen": 627638272 }, { "epoch": 0.19, "learning_rate": 0.00040900702106318957, "loss": 2.8026, "theoretical_loss": 3.821934879216985, "tokens_seen": 627703808 }, { "epoch": 0.19, "learning_rate": 0.00040899699097291875, "loss": 2.8184, "theoretical_loss": 3.8218935737487527, "tokens_seen": 627769344 }, { "epoch": 0.19, "learning_rate": 0.000408986960882648, "loss": 2.7926, "theoretical_loss": 3.821852273799621, "tokens_seen": 627834880 }, { "epoch": 0.19, "learning_rate": 0.0004089769307923771, "loss": 2.8435, "theoretical_loss": 3.8218109793682773, "tokens_seen": 627900416 }, { "epoch": 0.19, "learning_rate": 0.00040896690070210635, "loss": 2.7916, "theoretical_loss": 3.821769690453407, "tokens_seen": 627965952 }, { "epoch": 0.19, "learning_rate": 0.00040895687061183553, "loss": 3.0103, "theoretical_loss": 3.8217284070536994, "tokens_seen": 628031488 }, { "epoch": 0.19, "learning_rate": 0.0004089468405215647, "loss": 2.9298, "theoretical_loss": 3.8216871291678416, "tokens_seen": 628097024 }, { "epoch": 0.19, "learning_rate": 0.0004089368104312939, "loss": 2.8847, "theoretical_loss": 3.8216458567945217, "tokens_seen": 628162560 }, { "epoch": 0.19, "learning_rate": 0.0004089267803410231, "loss": 2.7306, "theoretical_loss": 3.821604589932429, "tokens_seen": 628228096 }, { "epoch": 0.19, "learning_rate": 0.00040891675025075226, "loss": 3.0658, "theoretical_loss": 3.821563328580253, "tokens_seen": 628293632 }, { "epoch": 0.19, "learning_rate": 0.0004089067201604815, "loss": 2.9798, "theoretical_loss": 3.8215220727366823, "tokens_seen": 628359168 }, { "epoch": 0.19, "learning_rate": 0.0004088966900702106, "loss": 2.7955, "theoretical_loss": 3.8214808224004075, "tokens_seen": 628424704 }, { "epoch": 0.19, "learning_rate": 0.00040888665997993985, "loss": 2.8123, "theoretical_loss": 3.82143957757012, "tokens_seen": 628490240 }, { "epoch": 0.19, "learning_rate": 0.000408876629889669, "loss": 2.7577, "theoretical_loss": 3.82139833824451, "tokens_seen": 628555776 }, { "epoch": 0.19, "learning_rate": 0.0004088665997993982, "loss": 2.6647, "theoretical_loss": 3.8213571044222694, "tokens_seen": 628621312 }, { "epoch": 0.19, "learning_rate": 0.0004088565697091274, "loss": 2.8947, "theoretical_loss": 3.82131587610209, "tokens_seen": 628686848 }, { "epoch": 0.19, "learning_rate": 0.0004088465396188566, "loss": 2.7107, "theoretical_loss": 3.8212746532826642, "tokens_seen": 628752384 }, { "epoch": 0.19, "learning_rate": 0.00040883650952858576, "loss": 2.9835, "theoretical_loss": 3.821233435962685, "tokens_seen": 628817920 }, { "epoch": 0.19, "learning_rate": 0.000408826479438315, "loss": 2.8205, "theoretical_loss": 3.8211922241408454, "tokens_seen": 628883456 }, { "epoch": 0.19, "learning_rate": 0.0004088164493480441, "loss": 2.8666, "theoretical_loss": 3.821151017815839, "tokens_seen": 628948992 }, { "epoch": 0.19, "learning_rate": 0.00040880641925777336, "loss": 2.8553, "theoretical_loss": 3.8211098169863607, "tokens_seen": 629014528 }, { "epoch": 0.19, "learning_rate": 0.0004087963891675025, "loss": 2.689, "theoretical_loss": 3.8210686216511043, "tokens_seen": 629080064 }, { "epoch": 0.19, "objective/train/docs_used": 365375, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6356372833251953, "objective/train/theoretical_loss": 3.8210274318087656, "objective/train/tokens_used": 649605600, "theoretical_loss": 3.8210274318087656, "tokens_seen": 629145600 }, { "epoch": 0.19, "learning_rate": 0.0004087863590772317, "loss": 2.6537, "theoretical_loss": 3.8210274318087656, "tokens_seen": 629145600 }, { "epoch": 0.19, "learning_rate": 0.0004087763289869609, "loss": 2.9341, "theoretical_loss": 3.8209862474580394, "tokens_seen": 629211136 }, { "epoch": 0.19, "learning_rate": 0.0004087662988966901, "loss": 2.9428, "theoretical_loss": 3.8209450685976227, "tokens_seen": 629276672 }, { "epoch": 0.19, "learning_rate": 0.00040875626880641926, "loss": 2.7736, "theoretical_loss": 3.820903895226211, "tokens_seen": 629342208 }, { "epoch": 0.19, "learning_rate": 0.00040874623871614844, "loss": 2.9722, "theoretical_loss": 3.820862727342501, "tokens_seen": 629407744 }, { "epoch": 0.19, "learning_rate": 0.0004087362086258776, "loss": 3.2209, "theoretical_loss": 3.8208215649451907, "tokens_seen": 629473280 }, { "epoch": 0.19, "learning_rate": 0.00040872617853560686, "loss": 3.0297, "theoretical_loss": 3.8207804080329772, "tokens_seen": 629538816 }, { "epoch": 0.19, "learning_rate": 0.000408716148445336, "loss": 2.4412, "theoretical_loss": 3.8207392566045595, "tokens_seen": 629604352 }, { "epoch": 0.19, "learning_rate": 0.0004087061183550652, "loss": 2.7719, "theoretical_loss": 3.8206981106586357, "tokens_seen": 629669888 }, { "epoch": 0.19, "learning_rate": 0.00040869608826479435, "loss": 2.9634, "theoretical_loss": 3.820656970193905, "tokens_seen": 629735424 }, { "epoch": 0.19, "learning_rate": 0.0004086860581745236, "loss": 3.0284, "theoretical_loss": 3.8206158352090664, "tokens_seen": 629800960 }, { "epoch": 0.19, "learning_rate": 0.00040867602808425277, "loss": 2.9308, "theoretical_loss": 3.820574705702821, "tokens_seen": 629866496 }, { "epoch": 0.19, "learning_rate": 0.00040866599799398195, "loss": 2.8229, "theoretical_loss": 3.8205335816738684, "tokens_seen": 629932032 }, { "epoch": 0.19, "learning_rate": 0.00040865596790371113, "loss": 2.858, "theoretical_loss": 3.8204924631209094, "tokens_seen": 629997568 }, { "epoch": 0.19, "learning_rate": 0.00040864593781344036, "loss": 2.7671, "theoretical_loss": 3.820451350042646, "tokens_seen": 630063104 }, { "epoch": 0.19, "learning_rate": 0.0004086359077231695, "loss": 2.662, "theoretical_loss": 3.820410242437779, "tokens_seen": 630128640 }, { "epoch": 0.19, "learning_rate": 0.0004086258776328987, "loss": 3.0353, "theoretical_loss": 3.8203691403050106, "tokens_seen": 630194176 }, { "epoch": 0.19, "learning_rate": 0.00040861584754262785, "loss": 2.8499, "theoretical_loss": 3.8203280436430447, "tokens_seen": 630259712 }, { "epoch": 0.19, "learning_rate": 0.0004086058174523571, "loss": 2.6793, "theoretical_loss": 3.8202869524505827, "tokens_seen": 630325248 }, { "epoch": 0.19, "learning_rate": 0.00040859578736208627, "loss": 2.764, "theoretical_loss": 3.8202458667263297, "tokens_seen": 630390784 }, { "epoch": 0.19, "learning_rate": 0.00040858575727181545, "loss": 3.1935, "theoretical_loss": 3.8202047864689885, "tokens_seen": 630456320 }, { "epoch": 0.19, "learning_rate": 0.00040857572718154463, "loss": 3.0614, "theoretical_loss": 3.8201637116772638, "tokens_seen": 630521856 }, { "epoch": 0.19, "learning_rate": 0.0004085656970912738, "loss": 2.7295, "theoretical_loss": 3.820122642349861, "tokens_seen": 630587392 }, { "epoch": 0.19, "learning_rate": 0.000408555667001003, "loss": 2.9255, "theoretical_loss": 3.8200815784854845, "tokens_seen": 630652928 }, { "epoch": 0.19, "learning_rate": 0.00040854563691073223, "loss": 2.8853, "theoretical_loss": 3.8200405200828405, "tokens_seen": 630718464 }, { "epoch": 0.19, "objective/train/docs_used": 366612, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8384101390838623, "objective/train/theoretical_loss": 3.8199994671406348, "objective/train/tokens_used": 651244000, "theoretical_loss": 3.8199994671406348, "tokens_seen": 630784000 }, { "epoch": 0.19, "learning_rate": 0.00040853560682046136, "loss": 2.7613, "theoretical_loss": 3.8199994671406348, "tokens_seen": 630784000 }, { "epoch": 0.19, "learning_rate": 0.0004085255767301906, "loss": 2.9038, "theoretical_loss": 3.819958419657574, "tokens_seen": 630849536 }, { "epoch": 0.19, "learning_rate": 0.0004085155466399197, "loss": 2.9814, "theoretical_loss": 3.8199173776323665, "tokens_seen": 630915072 }, { "epoch": 0.19, "learning_rate": 0.00040850551654964895, "loss": 2.8793, "theoretical_loss": 3.819876341063718, "tokens_seen": 630980608 }, { "epoch": 0.19, "learning_rate": 0.00040849548645937814, "loss": 2.7772, "theoretical_loss": 3.8198353099503373, "tokens_seen": 631046144 }, { "epoch": 0.19, "learning_rate": 0.0004084854563691073, "loss": 2.9659, "theoretical_loss": 3.8197942842909325, "tokens_seen": 631111680 }, { "epoch": 0.19, "learning_rate": 0.0004084754262788365, "loss": 3.0307, "theoretical_loss": 3.8197532640842127, "tokens_seen": 631177216 }, { "epoch": 0.19, "learning_rate": 0.00040846539618856573, "loss": 3.0699, "theoretical_loss": 3.8197122493288864, "tokens_seen": 631242752 }, { "epoch": 0.19, "learning_rate": 0.00040845536609829486, "loss": 2.6885, "theoretical_loss": 3.819671240023664, "tokens_seen": 631308288 }, { "epoch": 0.19, "learning_rate": 0.0004084453360080241, "loss": 2.8712, "theoretical_loss": 3.8196302361672556, "tokens_seen": 631373824 }, { "epoch": 0.19, "learning_rate": 0.0004084353059177532, "loss": 2.7389, "theoretical_loss": 3.8195892377583713, "tokens_seen": 631439360 }, { "epoch": 0.19, "learning_rate": 0.00040842527582748246, "loss": 2.9399, "theoretical_loss": 3.8195482447957225, "tokens_seen": 631504896 }, { "epoch": 0.19, "learning_rate": 0.00040841524573721164, "loss": 2.932, "theoretical_loss": 3.819507257278021, "tokens_seen": 631570432 }, { "epoch": 0.19, "learning_rate": 0.0004084052156469408, "loss": 2.8941, "theoretical_loss": 3.8194662752039776, "tokens_seen": 631635968 }, { "epoch": 0.19, "learning_rate": 0.00040839518555667005, "loss": 2.5435, "theoretical_loss": 3.8194252985723054, "tokens_seen": 631701504 }, { "epoch": 0.19, "learning_rate": 0.0004083851554663992, "loss": 2.7923, "theoretical_loss": 3.8193843273817167, "tokens_seen": 631767040 }, { "epoch": 0.19, "learning_rate": 0.0004083751253761284, "loss": 2.8786, "theoretical_loss": 3.819343361630925, "tokens_seen": 631832576 }, { "epoch": 0.19, "learning_rate": 0.0004083650952858576, "loss": 2.6148, "theoretical_loss": 3.8193024013186445, "tokens_seen": 631898112 }, { "epoch": 0.19, "learning_rate": 0.0004083550651955868, "loss": 3.0115, "theoretical_loss": 3.819261446443588, "tokens_seen": 631963648 }, { "epoch": 0.19, "learning_rate": 0.00040834503510531596, "loss": 2.8173, "theoretical_loss": 3.819220497004471, "tokens_seen": 632029184 }, { "epoch": 0.19, "learning_rate": 0.0004083350050150452, "loss": 2.8355, "theoretical_loss": 3.8191795530000077, "tokens_seen": 632094720 }, { "epoch": 0.19, "learning_rate": 0.0004083249749247743, "loss": 3.0112, "theoretical_loss": 3.8191386144289137, "tokens_seen": 632160256 }, { "epoch": 0.19, "learning_rate": 0.00040831494483450356, "loss": 2.8569, "theoretical_loss": 3.8190976812899056, "tokens_seen": 632225792 }, { "epoch": 0.19, "learning_rate": 0.0004083049147442327, "loss": 3.1, "theoretical_loss": 3.819056753581699, "tokens_seen": 632291328 }, { "epoch": 0.19, "learning_rate": 0.0004082948846539619, "loss": 2.8298, "theoretical_loss": 3.8190158313030103, "tokens_seen": 632356864 }, { "epoch": 0.19, "objective/train/docs_used": 367323, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.4437427520751953, "objective/train/theoretical_loss": 3.818974914452557, "objective/train/tokens_used": 652882400, "theoretical_loss": 3.818974914452557, "tokens_seen": 632422400 }, { "epoch": 0.19, "learning_rate": 0.0004082848545636911, "loss": 3.038, "theoretical_loss": 3.818974914452557, "tokens_seen": 632422400 }, { "epoch": 0.19, "learning_rate": 0.0004082748244734203, "loss": 2.9644, "theoretical_loss": 3.8189340030290566, "tokens_seen": 632487936 }, { "epoch": 0.19, "learning_rate": 0.00040826479438314946, "loss": 2.883, "theoretical_loss": 3.8188930970312267, "tokens_seen": 632553472 }, { "epoch": 0.19, "learning_rate": 0.00040825476429287864, "loss": 2.8525, "theoretical_loss": 3.8188521964577866, "tokens_seen": 632619008 }, { "epoch": 0.19, "learning_rate": 0.0004082447342026078, "loss": 3.0628, "theoretical_loss": 3.8188113013074547, "tokens_seen": 632684544 }, { "epoch": 0.19, "learning_rate": 0.00040823470411233706, "loss": 2.7063, "theoretical_loss": 3.8187704115789503, "tokens_seen": 632750080 }, { "epoch": 0.19, "learning_rate": 0.0004082246740220662, "loss": 2.9827, "theoretical_loss": 3.8187295272709934, "tokens_seen": 632815616 }, { "epoch": 0.19, "learning_rate": 0.0004082146439317954, "loss": 2.714, "theoretical_loss": 3.8186886483823033, "tokens_seen": 632881152 }, { "epoch": 0.19, "learning_rate": 0.00040820461384152455, "loss": 3.1194, "theoretical_loss": 3.818647774911601, "tokens_seen": 632946688 }, { "epoch": 0.19, "learning_rate": 0.0004081945837512538, "loss": 2.8226, "theoretical_loss": 3.8186069068576085, "tokens_seen": 633012224 }, { "epoch": 0.19, "learning_rate": 0.00040818455366098297, "loss": 2.7235, "theoretical_loss": 3.818566044219046, "tokens_seen": 633077760 }, { "epoch": 0.19, "learning_rate": 0.00040817452357071215, "loss": 2.9174, "theoretical_loss": 3.818525186994636, "tokens_seen": 633143296 }, { "epoch": 0.19, "learning_rate": 0.00040816449348044133, "loss": 2.9005, "theoretical_loss": 3.8184843351831006, "tokens_seen": 633208832 }, { "epoch": 0.19, "learning_rate": 0.00040815446339017056, "loss": 2.8442, "theoretical_loss": 3.8184434887831626, "tokens_seen": 633274368 }, { "epoch": 0.19, "learning_rate": 0.0004081444332998997, "loss": 2.8236, "theoretical_loss": 3.8184026477935458, "tokens_seen": 633339904 }, { "epoch": 0.19, "learning_rate": 0.0004081344032096289, "loss": 2.5517, "theoretical_loss": 3.8183618122129737, "tokens_seen": 633405440 }, { "epoch": 0.19, "learning_rate": 0.00040812437311935805, "loss": 3.1487, "theoretical_loss": 3.818320982040169, "tokens_seen": 633470976 }, { "epoch": 0.19, "learning_rate": 0.0004081143430290873, "loss": 2.7316, "theoretical_loss": 3.8182801572738585, "tokens_seen": 633536512 }, { "epoch": 0.19, "learning_rate": 0.00040810431293881647, "loss": 3.1412, "theoretical_loss": 3.8182393379127655, "tokens_seen": 633602048 }, { "epoch": 0.19, "learning_rate": 0.00040809428284854565, "loss": 2.8875, "theoretical_loss": 3.8181985239556155, "tokens_seen": 633667584 }, { "epoch": 0.19, "learning_rate": 0.00040808425275827483, "loss": 3.026, "theoretical_loss": 3.8181577154011346, "tokens_seen": 633733120 }, { "epoch": 0.19, "learning_rate": 0.000408074222668004, "loss": 2.8875, "theoretical_loss": 3.8181169122480494, "tokens_seen": 633798656 }, { "epoch": 0.19, "learning_rate": 0.0004080641925777332, "loss": 2.882, "theoretical_loss": 3.818076114495086, "tokens_seen": 633864192 }, { "epoch": 0.19, "learning_rate": 0.00040805416248746243, "loss": 2.8508, "theoretical_loss": 3.8180353221409717, "tokens_seen": 633929728 }, { "epoch": 0.19, "learning_rate": 0.00040804413239719156, "loss": 2.892, "theoretical_loss": 3.817994535184434, "tokens_seen": 633995264 }, { "epoch": 0.19, "objective/train/docs_used": 367883, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8706178665161133, "objective/train/theoretical_loss": 3.8179537536242005, "objective/train/tokens_used": 654520800, "theoretical_loss": 3.8179537536242005, "tokens_seen": 634060800 }, { "epoch": 0.19, "learning_rate": 0.0004080341023069208, "loss": 2.9069, "theoretical_loss": 3.8179537536242005, "tokens_seen": 634060800 }, { "epoch": 0.19, "learning_rate": 0.0004080240722166499, "loss": 2.9371, "theoretical_loss": 3.8179129774590006, "tokens_seen": 634126336 }, { "epoch": 0.19, "learning_rate": 0.00040801404212637915, "loss": 2.7872, "theoretical_loss": 3.8178722066875626, "tokens_seen": 634191872 }, { "epoch": 0.19, "learning_rate": 0.00040800401203610834, "loss": 2.8084, "theoretical_loss": 3.8178314413086154, "tokens_seen": 634257408 }, { "epoch": 0.19, "learning_rate": 0.0004079939819458375, "loss": 2.8125, "theoretical_loss": 3.817790681320889, "tokens_seen": 634322944 }, { "epoch": 0.19, "learning_rate": 0.0004079839518555667, "loss": 2.9918, "theoretical_loss": 3.817749926723114, "tokens_seen": 634388480 }, { "epoch": 0.19, "learning_rate": 0.00040797392176529593, "loss": 3.0913, "theoretical_loss": 3.8177091775140197, "tokens_seen": 634454016 }, { "epoch": 0.19, "learning_rate": 0.00040796389167502506, "loss": 3.0007, "theoretical_loss": 3.8176684336923383, "tokens_seen": 634519552 }, { "epoch": 0.19, "learning_rate": 0.0004079538615847543, "loss": 2.9475, "theoretical_loss": 3.8176276952568005, "tokens_seen": 634585088 }, { "epoch": 0.19, "learning_rate": 0.0004079438314944834, "loss": 2.7219, "theoretical_loss": 3.8175869622061382, "tokens_seen": 634650624 }, { "epoch": 0.19, "learning_rate": 0.00040793380140421266, "loss": 2.8263, "theoretical_loss": 3.8175462345390843, "tokens_seen": 634716160 }, { "epoch": 0.19, "learning_rate": 0.00040792377131394184, "loss": 2.8433, "theoretical_loss": 3.817505512254371, "tokens_seen": 634781696 }, { "epoch": 0.19, "learning_rate": 0.000407913741223671, "loss": 2.7156, "theoretical_loss": 3.8174647953507312, "tokens_seen": 634847232 }, { "epoch": 0.19, "learning_rate": 0.0004079037111334002, "loss": 2.9948, "theoretical_loss": 3.817424083826899, "tokens_seen": 634912768 }, { "epoch": 0.19, "learning_rate": 0.0004078936810431294, "loss": 2.8635, "theoretical_loss": 3.817383377681608, "tokens_seen": 634978304 }, { "epoch": 0.19, "learning_rate": 0.00040788365095285856, "loss": 2.8047, "theoretical_loss": 3.8173426769135927, "tokens_seen": 635043840 }, { "epoch": 0.19, "learning_rate": 0.0004078736208625878, "loss": 2.8993, "theoretical_loss": 3.8173019815215885, "tokens_seen": 635109376 }, { "epoch": 0.19, "learning_rate": 0.0004078635907723169, "loss": 2.8109, "theoretical_loss": 3.8172612915043294, "tokens_seen": 635174912 }, { "epoch": 0.19, "learning_rate": 0.00040785356068204616, "loss": 2.8459, "theoretical_loss": 3.8172206068605523, "tokens_seen": 635240448 }, { "epoch": 0.19, "learning_rate": 0.0004078435305917753, "loss": 2.9693, "theoretical_loss": 3.8171799275889926, "tokens_seen": 635305984 }, { "epoch": 0.19, "learning_rate": 0.0004078335005015045, "loss": 2.8409, "theoretical_loss": 3.8171392536883877, "tokens_seen": 635371520 }, { "epoch": 0.19, "learning_rate": 0.0004078234704112337, "loss": 2.6479, "theoretical_loss": 3.8170985851574732, "tokens_seen": 635437056 }, { "epoch": 0.19, "learning_rate": 0.0004078134403209629, "loss": 2.7721, "theoretical_loss": 3.8170579219949876, "tokens_seen": 635502592 }, { "epoch": 0.19, "learning_rate": 0.00040780341023069207, "loss": 2.959, "theoretical_loss": 3.8170172641996682, "tokens_seen": 635568128 }, { "epoch": 0.19, "learning_rate": 0.0004077933801404213, "loss": 2.8975, "theoretical_loss": 3.816976611770254, "tokens_seen": 635633664 }, { "epoch": 0.19, "objective/train/docs_used": 369288, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.124577522277832, "objective/train/theoretical_loss": 3.8169359647054835, "objective/train/tokens_used": 656159200, "theoretical_loss": 3.8169359647054835, "tokens_seen": 635699200 }, { "epoch": 0.19, "learning_rate": 0.00040778335005015043, "loss": 2.8436, "theoretical_loss": 3.8169359647054835, "tokens_seen": 635699200 }, { "epoch": 0.19, "learning_rate": 0.00040777331995987966, "loss": 2.9538, "theoretical_loss": 3.8168953230040947, "tokens_seen": 635764736 }, { "epoch": 0.19, "learning_rate": 0.0004077632898696088, "loss": 2.9674, "theoretical_loss": 3.8168546866648283, "tokens_seen": 635830272 }, { "epoch": 0.19, "learning_rate": 0.000407753259779338, "loss": 3.0346, "theoretical_loss": 3.816814055686424, "tokens_seen": 635895808 }, { "epoch": 0.19, "learning_rate": 0.0004077432296890672, "loss": 2.922, "theoretical_loss": 3.816773430067622, "tokens_seen": 635961344 }, { "epoch": 0.19, "learning_rate": 0.0004077331995987964, "loss": 2.7976, "theoretical_loss": 3.816732809807163, "tokens_seen": 636026880 }, { "epoch": 0.19, "learning_rate": 0.00040772316950852557, "loss": 2.849, "theoretical_loss": 3.816692194903789, "tokens_seen": 636092416 }, { "epoch": 0.19, "learning_rate": 0.00040771313941825475, "loss": 2.9071, "theoretical_loss": 3.816651585356241, "tokens_seen": 636157952 }, { "epoch": 0.19, "learning_rate": 0.00040770310932798393, "loss": 2.7778, "theoretical_loss": 3.816610981163261, "tokens_seen": 636223488 }, { "epoch": 0.19, "learning_rate": 0.00040769307923771317, "loss": 2.7555, "theoretical_loss": 3.816570382323592, "tokens_seen": 636289024 }, { "epoch": 0.19, "learning_rate": 0.0004076830491474423, "loss": 2.969, "theoretical_loss": 3.8165297888359766, "tokens_seen": 636354560 }, { "epoch": 0.19, "learning_rate": 0.00040767301905717153, "loss": 2.778, "theoretical_loss": 3.8164892006991584, "tokens_seen": 636420096 }, { "epoch": 0.19, "learning_rate": 0.0004076629889669007, "loss": 3.004, "theoretical_loss": 3.816448617911881, "tokens_seen": 636485632 }, { "epoch": 0.19, "learning_rate": 0.0004076529588766299, "loss": 2.911, "theoretical_loss": 3.816408040472889, "tokens_seen": 636551168 }, { "epoch": 0.19, "learning_rate": 0.0004076429287863591, "loss": 2.8413, "theoretical_loss": 3.816367468380926, "tokens_seen": 636616704 }, { "epoch": 0.19, "learning_rate": 0.00040763289869608825, "loss": 2.788, "theoretical_loss": 3.8163269016347385, "tokens_seen": 636682240 }, { "epoch": 0.19, "learning_rate": 0.0004076228686058175, "loss": 3.115, "theoretical_loss": 3.8162863402330713, "tokens_seen": 636747776 }, { "epoch": 0.19, "learning_rate": 0.00040761283851554667, "loss": 2.6516, "theoretical_loss": 3.81624578417467, "tokens_seen": 636813312 }, { "epoch": 0.19, "learning_rate": 0.00040760280842527585, "loss": 2.9217, "theoretical_loss": 3.816205233458281, "tokens_seen": 636878848 }, { "epoch": 0.19, "learning_rate": 0.00040759277833500503, "loss": 2.7958, "theoretical_loss": 3.816164688082652, "tokens_seen": 636944384 }, { "epoch": 0.19, "learning_rate": 0.0004075827482447342, "loss": 2.5185, "theoretical_loss": 3.816124148046529, "tokens_seen": 637009920 }, { "epoch": 0.19, "learning_rate": 0.0004075727181544634, "loss": 3.0614, "theoretical_loss": 3.81608361334866, "tokens_seen": 637075456 }, { "epoch": 0.19, "learning_rate": 0.00040756268806419263, "loss": 2.9743, "theoretical_loss": 3.8160430839877937, "tokens_seen": 637140992 }, { "epoch": 0.19, "learning_rate": 0.00040755265797392176, "loss": 2.926, "theoretical_loss": 3.8160025599626777, "tokens_seen": 637206528 }, { "epoch": 0.19, "learning_rate": 0.000407542627883651, "loss": 2.7422, "theoretical_loss": 3.815962041272061, "tokens_seen": 637272064 }, { "epoch": 0.19, "objective/train/docs_used": 369983, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8315579891204834, "objective/train/theoretical_loss": 3.815921527914693, "objective/train/tokens_used": 657797600, "theoretical_loss": 3.815921527914693, "tokens_seen": 637337600 }, { "epoch": 0.19, "learning_rate": 0.0004075325977933801, "loss": 2.9318, "theoretical_loss": 3.815921527914693, "tokens_seen": 637337600 }, { "epoch": 0.19, "learning_rate": 0.00040752256770310935, "loss": 2.6521, "theoretical_loss": 3.815881019889324, "tokens_seen": 637403136 }, { "epoch": 0.19, "learning_rate": 0.00040751253761283854, "loss": 2.8127, "theoretical_loss": 3.8158405171947027, "tokens_seen": 637468672 }, { "epoch": 0.19, "learning_rate": 0.0004075025075225677, "loss": 3.033, "theoretical_loss": 3.8158000198295814, "tokens_seen": 637534208 }, { "epoch": 0.19, "learning_rate": 0.0004074924774322969, "loss": 3.1116, "theoretical_loss": 3.81575952779271, "tokens_seen": 637599744 }, { "epoch": 0.19, "learning_rate": 0.00040748244734202613, "loss": 2.9631, "theoretical_loss": 3.81571904108284, "tokens_seen": 637665280 }, { "epoch": 0.19, "learning_rate": 0.00040747241725175526, "loss": 2.7844, "theoretical_loss": 3.815678559698723, "tokens_seen": 637730816 }, { "epoch": 0.19, "learning_rate": 0.0004074623871614845, "loss": 2.7158, "theoretical_loss": 3.8156380836391124, "tokens_seen": 637796352 }, { "epoch": 0.19, "learning_rate": 0.0004074523570712136, "loss": 2.7155, "theoretical_loss": 3.8155976129027596, "tokens_seen": 637861888 }, { "epoch": 0.19, "learning_rate": 0.00040744232698094286, "loss": 3.0124, "theoretical_loss": 3.815557147488418, "tokens_seen": 637927424 }, { "epoch": 0.19, "learning_rate": 0.00040743229689067204, "loss": 2.8726, "theoretical_loss": 3.8155166873948416, "tokens_seen": 637992960 }, { "epoch": 0.19, "learning_rate": 0.0004074222668004012, "loss": 3.0371, "theoretical_loss": 3.815476232620784, "tokens_seen": 638058496 }, { "epoch": 0.19, "learning_rate": 0.0004074122367101304, "loss": 2.7492, "theoretical_loss": 3.8154357831649994, "tokens_seen": 638124032 }, { "epoch": 0.19, "learning_rate": 0.0004074022066198596, "loss": 2.8076, "theoretical_loss": 3.8153953390262427, "tokens_seen": 638189568 }, { "epoch": 0.19, "learning_rate": 0.00040739217652958876, "loss": 2.772, "theoretical_loss": 3.815354900203269, "tokens_seen": 638255104 }, { "epoch": 0.19, "learning_rate": 0.000407382146439318, "loss": 2.7375, "theoretical_loss": 3.8153144666948338, "tokens_seen": 638320640 }, { "epoch": 0.19, "learning_rate": 0.0004073721163490471, "loss": 2.8696, "theoretical_loss": 3.8152740384996937, "tokens_seen": 638386176 }, { "epoch": 0.19, "learning_rate": 0.00040736208625877636, "loss": 2.7808, "theoretical_loss": 3.815233615616605, "tokens_seen": 638451712 }, { "epoch": 0.19, "learning_rate": 0.0004073520561685055, "loss": 2.7748, "theoretical_loss": 3.815193198044324, "tokens_seen": 638517248 }, { "epoch": 0.19, "learning_rate": 0.0004073420260782347, "loss": 2.7993, "theoretical_loss": 3.8151527857816077, "tokens_seen": 638582784 }, { "epoch": 0.19, "learning_rate": 0.0004073319959879639, "loss": 2.6915, "theoretical_loss": 3.815112378827215, "tokens_seen": 638648320 }, { "epoch": 0.19, "learning_rate": 0.0004073219658976931, "loss": 3.0531, "theoretical_loss": 3.815071977179903, "tokens_seen": 638713856 }, { "epoch": 0.19, "learning_rate": 0.00040731193580742227, "loss": 2.9624, "theoretical_loss": 3.815031580838431, "tokens_seen": 638779392 }, { "epoch": 0.19, "learning_rate": 0.0004073019057171515, "loss": 3.0607, "theoretical_loss": 3.814991189801557, "tokens_seen": 638844928 }, { "epoch": 0.19, "learning_rate": 0.00040729187562688063, "loss": 2.6183, "theoretical_loss": 3.8149508040680415, "tokens_seen": 638910464 }, { "epoch": 0.19, "objective/train/docs_used": 370882, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.928964853286743, "objective/train/theoretical_loss": 3.8149104236366433, "objective/train/tokens_used": 659436000, "theoretical_loss": 3.8149104236366433, "tokens_seen": 638976000 }, { "epoch": 0.19, "learning_rate": 0.00040728184553660986, "loss": 2.8016, "theoretical_loss": 3.8149104236366433, "tokens_seen": 638976000 }, { "epoch": 0.19, "learning_rate": 0.000407271815446339, "loss": 2.7656, "theoretical_loss": 3.8148700485061235, "tokens_seen": 639041536 }, { "epoch": 0.19, "learning_rate": 0.0004072617853560682, "loss": 2.8697, "theoretical_loss": 3.814829678675242, "tokens_seen": 639107072 }, { "epoch": 0.19, "learning_rate": 0.0004072517552657974, "loss": 2.848, "theoretical_loss": 3.81478931414276, "tokens_seen": 639172608 }, { "epoch": 0.19, "learning_rate": 0.0004072417251755266, "loss": 2.922, "theoretical_loss": 3.814748954907439, "tokens_seen": 639238144 }, { "epoch": 0.19, "learning_rate": 0.00040723169508525577, "loss": 3.021, "theoretical_loss": 3.814708600968041, "tokens_seen": 639303680 }, { "epoch": 0.19, "learning_rate": 0.00040722166499498495, "loss": 2.9457, "theoretical_loss": 3.8146682523233286, "tokens_seen": 639369216 }, { "epoch": 0.19, "learning_rate": 0.00040721163490471413, "loss": 2.967, "theoretical_loss": 3.8146279089720636, "tokens_seen": 639434752 }, { "epoch": 0.19, "learning_rate": 0.00040720160481444337, "loss": 3.1371, "theoretical_loss": 3.8145875709130093, "tokens_seen": 639500288 }, { "epoch": 0.19, "learning_rate": 0.0004071915747241725, "loss": 2.9063, "theoretical_loss": 3.8145472381449297, "tokens_seen": 639565824 }, { "epoch": 0.19, "learning_rate": 0.00040718154463390173, "loss": 2.898, "theoretical_loss": 3.814506910666589, "tokens_seen": 639631360 }, { "epoch": 0.19, "learning_rate": 0.0004071715145436309, "loss": 2.8287, "theoretical_loss": 3.814466588476751, "tokens_seen": 639696896 }, { "epoch": 0.19, "learning_rate": 0.0004071614844533601, "loss": 2.9777, "theoretical_loss": 3.81442627157418, "tokens_seen": 639762432 }, { "epoch": 0.19, "learning_rate": 0.00040715145436308927, "loss": 2.9235, "theoretical_loss": 3.814385959957643, "tokens_seen": 639827968 }, { "epoch": 0.19, "learning_rate": 0.00040714142427281845, "loss": 2.8904, "theoretical_loss": 3.8143456536259035, "tokens_seen": 639893504 }, { "epoch": 0.19, "learning_rate": 0.00040713139418254763, "loss": 2.8318, "theoretical_loss": 3.814305352577729, "tokens_seen": 639959040 }, { "epoch": 0.19, "learning_rate": 0.00040712136409227687, "loss": 3.0163, "theoretical_loss": 3.8142650568118848, "tokens_seen": 640024576 }, { "epoch": 0.19, "learning_rate": 0.000407111334002006, "loss": 2.8352, "theoretical_loss": 3.8142247663271385, "tokens_seen": 640090112 }, { "epoch": 0.19, "learning_rate": 0.00040710130391173523, "loss": 2.7951, "theoretical_loss": 3.814184481122257, "tokens_seen": 640155648 }, { "epoch": 0.19, "learning_rate": 0.00040709127382146436, "loss": 2.72, "theoretical_loss": 3.8141442011960085, "tokens_seen": 640221184 }, { "epoch": 0.19, "learning_rate": 0.0004070812437311936, "loss": 2.7045, "theoretical_loss": 3.814103926547161, "tokens_seen": 640286720 }, { "epoch": 0.19, "learning_rate": 0.0004070712136409228, "loss": 3.0761, "theoretical_loss": 3.8140636571744824, "tokens_seen": 640352256 }, { "epoch": 0.19, "learning_rate": 0.00040706118355065196, "loss": 2.9509, "theoretical_loss": 3.8140233930767424, "tokens_seen": 640417792 }, { "epoch": 0.19, "learning_rate": 0.00040705115346038114, "loss": 3.1579, "theoretical_loss": 3.8139831342527093, "tokens_seen": 640483328 }, { "epoch": 0.19, "learning_rate": 0.0004070411233701103, "loss": 2.7903, "theoretical_loss": 3.8139428807011537, "tokens_seen": 640548864 }, { "epoch": 0.19, "objective/train/docs_used": 371543, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0356507301330566, "objective/train/theoretical_loss": 3.8139026324208456, "objective/train/tokens_used": 661074400, "theoretical_loss": 3.8139026324208456, "tokens_seen": 640614400 }, { "epoch": 0.19, "learning_rate": 0.0004070310932798395, "loss": 2.8759, "theoretical_loss": 3.8139026324208456, "tokens_seen": 640614400 }, { "epoch": 0.19, "learning_rate": 0.00040702106318956874, "loss": 2.6784, "theoretical_loss": 3.8138623894105557, "tokens_seen": 640679936 }, { "epoch": 0.19, "learning_rate": 0.00040701103309929786, "loss": 2.9655, "theoretical_loss": 3.8138221516690547, "tokens_seen": 640745472 }, { "epoch": 0.19, "learning_rate": 0.0004070010030090271, "loss": 2.9818, "theoretical_loss": 3.813781919195114, "tokens_seen": 640811008 }, { "epoch": 0.19, "learning_rate": 0.0004069909729187563, "loss": 2.8611, "theoretical_loss": 3.813741691987506, "tokens_seen": 640876544 }, { "epoch": 0.19, "learning_rate": 0.00040698094282848546, "loss": 2.7041, "theoretical_loss": 3.813701470045001, "tokens_seen": 640942080 }, { "epoch": 0.19, "learning_rate": 0.00040697091273821464, "loss": 2.7796, "theoretical_loss": 3.8136612533663747, "tokens_seen": 641007616 }, { "epoch": 0.19, "learning_rate": 0.0004069608826479438, "loss": 2.6885, "theoretical_loss": 3.8136210419503977, "tokens_seen": 641073152 }, { "epoch": 0.19, "learning_rate": 0.000406950852557673, "loss": 3.197, "theoretical_loss": 3.813580835795844, "tokens_seen": 641138688 }, { "epoch": 0.19, "learning_rate": 0.00040694082246740224, "loss": 3.0585, "theoretical_loss": 3.8135406349014884, "tokens_seen": 641204224 }, { "epoch": 0.19, "learning_rate": 0.00040693079237713137, "loss": 3.0909, "theoretical_loss": 3.8135004392661047, "tokens_seen": 641269760 }, { "epoch": 0.19, "learning_rate": 0.0004069207622868606, "loss": 2.9761, "theoretical_loss": 3.8134602488884664, "tokens_seen": 641335296 }, { "epoch": 0.19, "learning_rate": 0.00040691073219658973, "loss": 2.8035, "theoretical_loss": 3.8134200637673503, "tokens_seen": 641400832 }, { "epoch": 0.19, "learning_rate": 0.00040690070210631896, "loss": 3.0324, "theoretical_loss": 3.813379883901531, "tokens_seen": 641466368 }, { "epoch": 0.19, "learning_rate": 0.0004068906720160482, "loss": 2.9224, "theoretical_loss": 3.8133397092897843, "tokens_seen": 641531904 }, { "epoch": 0.19, "learning_rate": 0.0004068806419257773, "loss": 2.6653, "theoretical_loss": 3.8132995399308873, "tokens_seen": 641597440 }, { "epoch": 0.19, "learning_rate": 0.00040687061183550656, "loss": 2.8266, "theoretical_loss": 3.813259375823616, "tokens_seen": 641662976 }, { "epoch": 0.19, "learning_rate": 0.0004068605817452357, "loss": 2.8396, "theoretical_loss": 3.8132192169667483, "tokens_seen": 641728512 }, { "epoch": 0.19, "learning_rate": 0.0004068505516549649, "loss": 3.1059, "theoretical_loss": 3.813179063359061, "tokens_seen": 641794048 }, { "epoch": 0.19, "learning_rate": 0.0004068405215646941, "loss": 2.8498, "theoretical_loss": 3.813138914999332, "tokens_seen": 641859584 }, { "epoch": 0.19, "learning_rate": 0.0004068304914744233, "loss": 2.9668, "theoretical_loss": 3.8130987718863407, "tokens_seen": 641925120 }, { "epoch": 0.19, "learning_rate": 0.00040682046138415247, "loss": 2.7385, "theoretical_loss": 3.813058634018865, "tokens_seen": 641990656 }, { "epoch": 0.19, "learning_rate": 0.0004068104312938817, "loss": 2.9445, "theoretical_loss": 3.8130185013956837, "tokens_seen": 642056192 }, { "epoch": 0.19, "learning_rate": 0.00040680040120361083, "loss": 3.2271, "theoretical_loss": 3.812978374015578, "tokens_seen": 642121728 }, { "epoch": 0.19, "learning_rate": 0.00040679037111334006, "loss": 3.0666, "theoretical_loss": 3.8129382518773265, "tokens_seen": 642187264 }, { "epoch": 0.19, "objective/train/docs_used": 373013, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.800529718399048, "objective/train/theoretical_loss": 3.8128981349797098, "objective/train/tokens_used": 662712800, "theoretical_loss": 3.8128981349797098, "tokens_seen": 642252800 }, { "epoch": 0.19, "learning_rate": 0.0004067803410230692, "loss": 2.8899, "theoretical_loss": 3.8128981349797098, "tokens_seen": 642252800 }, { "epoch": 0.19, "learning_rate": 0.0004067703109327984, "loss": 2.6683, "theoretical_loss": 3.8128580233215095, "tokens_seen": 642318336 }, { "epoch": 0.19, "learning_rate": 0.0004067602808425276, "loss": 2.9476, "theoretical_loss": 3.8128179169015057, "tokens_seen": 642383872 }, { "epoch": 0.19, "learning_rate": 0.0004067502507522568, "loss": 3.0143, "theoretical_loss": 3.8127778157184817, "tokens_seen": 642449408 }, { "epoch": 0.19, "learning_rate": 0.00040674022066198597, "loss": 2.8827, "theoretical_loss": 3.8127377197712176, "tokens_seen": 642514944 }, { "epoch": 0.19, "learning_rate": 0.00040673019057171515, "loss": 2.9197, "theoretical_loss": 3.8126976290584973, "tokens_seen": 642580480 }, { "epoch": 0.19, "learning_rate": 0.00040672016048144433, "loss": 2.6803, "theoretical_loss": 3.812657543579103, "tokens_seen": 642646016 }, { "epoch": 0.19, "learning_rate": 0.00040671013039117357, "loss": 2.8742, "theoretical_loss": 3.812617463331818, "tokens_seen": 642711552 }, { "epoch": 0.19, "learning_rate": 0.0004067001003009027, "loss": 2.8807, "theoretical_loss": 3.812577388315426, "tokens_seen": 642777088 }, { "epoch": 0.19, "learning_rate": 0.00040669007021063193, "loss": 3.1897, "theoretical_loss": 3.8125373185287117, "tokens_seen": 642842624 }, { "epoch": 0.19, "learning_rate": 0.0004066800401203611, "loss": 2.9624, "theoretical_loss": 3.812497253970459, "tokens_seen": 642908160 }, { "epoch": 0.19, "learning_rate": 0.0004066700100300903, "loss": 2.9739, "theoretical_loss": 3.8124571946394528, "tokens_seen": 642973696 }, { "epoch": 0.19, "learning_rate": 0.00040665997993981947, "loss": 2.8501, "theoretical_loss": 3.812417140534478, "tokens_seen": 643039232 }, { "epoch": 0.19, "learning_rate": 0.00040664994984954865, "loss": 2.9424, "theoretical_loss": 3.8123770916543216, "tokens_seen": 643104768 }, { "epoch": 0.19, "learning_rate": 0.00040663991975927783, "loss": 2.8588, "theoretical_loss": 3.8123370479977687, "tokens_seen": 643170304 }, { "epoch": 0.19, "learning_rate": 0.00040662988966900707, "loss": 3.0731, "theoretical_loss": 3.812297009563606, "tokens_seen": 643235840 }, { "epoch": 0.19, "learning_rate": 0.0004066198595787362, "loss": 3.135, "theoretical_loss": 3.8122569763506204, "tokens_seen": 643301376 }, { "epoch": 0.19, "learning_rate": 0.00040660982948846543, "loss": 2.7244, "theoretical_loss": 3.8122169483576, "tokens_seen": 643366912 }, { "epoch": 0.19, "learning_rate": 0.00040659979939819456, "loss": 2.6626, "theoretical_loss": 3.812176925583331, "tokens_seen": 643432448 }, { "epoch": 0.19, "learning_rate": 0.0004065897693079238, "loss": 2.9799, "theoretical_loss": 3.812136908026603, "tokens_seen": 643497984 }, { "epoch": 0.2, "learning_rate": 0.000406579739217653, "loss": 2.9877, "theoretical_loss": 3.8120968956862034, "tokens_seen": 643563520 }, { "epoch": 0.2, "learning_rate": 0.00040656970912738216, "loss": 2.865, "theoretical_loss": 3.8120568885609223, "tokens_seen": 643629056 }, { "epoch": 0.2, "learning_rate": 0.00040655967903711134, "loss": 2.8761, "theoretical_loss": 3.8120168866495483, "tokens_seen": 643694592 }, { "epoch": 0.2, "learning_rate": 0.0004065496489468405, "loss": 2.7782, "theoretical_loss": 3.8119768899508713, "tokens_seen": 643760128 }, { "epoch": 0.2, "learning_rate": 0.0004065396188565697, "loss": 2.9398, "theoretical_loss": 3.8119368984636814, "tokens_seen": 643825664 }, { "epoch": 0.2, "objective/train/docs_used": 373533, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.063260078430176, "objective/train/theoretical_loss": 3.8118969121867696, "objective/train/tokens_used": 664351200, "theoretical_loss": 3.8118969121867696, "tokens_seen": 643891200 }, { "epoch": 0.2, "learning_rate": 0.00040652958876629894, "loss": 2.9474, "theoretical_loss": 3.8118969121867696, "tokens_seen": 643891200 }, { "epoch": 0.2, "learning_rate": 0.00040651955867602806, "loss": 3.0836, "theoretical_loss": 3.8118569311189265, "tokens_seen": 643956736 }, { "epoch": 0.2, "learning_rate": 0.0004065095285857573, "loss": 2.8049, "theoretical_loss": 3.8118169552589434, "tokens_seen": 644022272 }, { "epoch": 0.2, "learning_rate": 0.0004064994984954865, "loss": 2.8489, "theoretical_loss": 3.811776984605612, "tokens_seen": 644087808 }, { "epoch": 0.2, "learning_rate": 0.00040648946840521566, "loss": 2.953, "theoretical_loss": 3.811737019157725, "tokens_seen": 644153344 }, { "epoch": 0.2, "learning_rate": 0.00040647943831494484, "loss": 3.0545, "theoretical_loss": 3.811697058914075, "tokens_seen": 644218880 }, { "epoch": 0.2, "learning_rate": 0.000406469408224674, "loss": 2.8362, "theoretical_loss": 3.8116571038734546, "tokens_seen": 644284416 }, { "epoch": 0.2, "learning_rate": 0.0004064593781344032, "loss": 2.8523, "theoretical_loss": 3.811617154034657, "tokens_seen": 644349952 }, { "epoch": 0.2, "learning_rate": 0.00040644934804413244, "loss": 2.8711, "theoretical_loss": 3.811577209396477, "tokens_seen": 644415488 }, { "epoch": 0.2, "learning_rate": 0.00040643931795386157, "loss": 2.8579, "theoretical_loss": 3.8115372699577073, "tokens_seen": 644481024 }, { "epoch": 0.2, "learning_rate": 0.0004064292878635908, "loss": 3.2256, "theoretical_loss": 3.8114973357171436, "tokens_seen": 644546560 }, { "epoch": 0.2, "learning_rate": 0.00040641925777331993, "loss": 2.8823, "theoretical_loss": 3.8114574066735805, "tokens_seen": 644612096 }, { "epoch": 0.2, "learning_rate": 0.00040640922768304916, "loss": 2.837, "theoretical_loss": 3.8114174828258136, "tokens_seen": 644677632 }, { "epoch": 0.2, "learning_rate": 0.00040639919759277834, "loss": 3.0595, "theoretical_loss": 3.8113775641726386, "tokens_seen": 644743168 }, { "epoch": 0.2, "learning_rate": 0.0004063891675025075, "loss": 2.956, "theoretical_loss": 3.8113376507128516, "tokens_seen": 644808704 }, { "epoch": 0.2, "learning_rate": 0.0004063791374122367, "loss": 2.7663, "theoretical_loss": 3.81129774244525, "tokens_seen": 644874240 }, { "epoch": 0.2, "learning_rate": 0.0004063691073219659, "loss": 2.9184, "theoretical_loss": 3.811257839368629, "tokens_seen": 644939776 }, { "epoch": 0.2, "learning_rate": 0.00040635907723169507, "loss": 2.7577, "theoretical_loss": 3.8112179414817877, "tokens_seen": 645005312 }, { "epoch": 0.2, "learning_rate": 0.0004063490471414243, "loss": 2.9307, "theoretical_loss": 3.811178048783524, "tokens_seen": 645070848 }, { "epoch": 0.2, "learning_rate": 0.00040633901705115343, "loss": 2.9966, "theoretical_loss": 3.8111381612726345, "tokens_seen": 645136384 }, { "epoch": 0.2, "learning_rate": 0.00040632898696088267, "loss": 3.0551, "theoretical_loss": 3.8110982789479193, "tokens_seen": 645201920 }, { "epoch": 0.2, "learning_rate": 0.00040631895687061185, "loss": 2.8291, "theoretical_loss": 3.811058401808177, "tokens_seen": 645267456 }, { "epoch": 0.2, "learning_rate": 0.00040630892678034103, "loss": 2.9193, "theoretical_loss": 3.8110185298522063, "tokens_seen": 645332992 }, { "epoch": 0.2, "learning_rate": 0.0004062988966900702, "loss": 2.9016, "theoretical_loss": 3.8109786630788083, "tokens_seen": 645398528 }, { "epoch": 0.2, "learning_rate": 0.0004062888665997994, "loss": 3.1083, "theoretical_loss": 3.8109388014867824, "tokens_seen": 645464064 }, { "epoch": 0.2, "objective/train/docs_used": 374553, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.623807430267334, "objective/train/theoretical_loss": 3.8108989450749293, "objective/train/tokens_used": 665989600, "theoretical_loss": 3.8108989450749293, "tokens_seen": 645529600 }, { "epoch": 0.2, "learning_rate": 0.00040627883650952857, "loss": 2.8018, "theoretical_loss": 3.8108989450749293, "tokens_seen": 645529600 }, { "epoch": 0.2, "learning_rate": 0.0004062688064192578, "loss": 2.8829, "theoretical_loss": 3.81085909384205, "tokens_seen": 645595136 }, { "epoch": 0.2, "learning_rate": 0.00040625877632898693, "loss": 2.9422, "theoretical_loss": 3.810819247786946, "tokens_seen": 645660672 }, { "epoch": 0.2, "learning_rate": 0.00040624874623871617, "loss": 2.9232, "theoretical_loss": 3.810779406908419, "tokens_seen": 645726208 }, { "epoch": 0.2, "learning_rate": 0.0004062387161484453, "loss": 2.9929, "theoretical_loss": 3.8107395712052714, "tokens_seen": 645791744 }, { "epoch": 0.2, "learning_rate": 0.00040622868605817453, "loss": 2.9012, "theoretical_loss": 3.810699740676306, "tokens_seen": 645857280 }, { "epoch": 0.2, "learning_rate": 0.0004062186559679037, "loss": 2.7108, "theoretical_loss": 3.8106599153203256, "tokens_seen": 645922816 }, { "epoch": 0.2, "learning_rate": 0.0004062086258776329, "loss": 2.9696, "theoretical_loss": 3.810620095136133, "tokens_seen": 645988352 }, { "epoch": 0.2, "learning_rate": 0.0004061985957873621, "loss": 2.7225, "theoretical_loss": 3.810580280122533, "tokens_seen": 646053888 }, { "epoch": 0.2, "learning_rate": 0.00040618856569709126, "loss": 3.0348, "theoretical_loss": 3.8105404702783288, "tokens_seen": 646119424 }, { "epoch": 0.2, "learning_rate": 0.00040617853560682044, "loss": 3.0608, "theoretical_loss": 3.8105006656023264, "tokens_seen": 646184960 }, { "epoch": 0.2, "learning_rate": 0.00040616850551654967, "loss": 2.9151, "theoretical_loss": 3.810460866093329, "tokens_seen": 646250496 }, { "epoch": 0.2, "learning_rate": 0.0004061584754262788, "loss": 2.8537, "theoretical_loss": 3.8104210717501434, "tokens_seen": 646316032 }, { "epoch": 0.2, "learning_rate": 0.00040614844533600804, "loss": 2.6747, "theoretical_loss": 3.8103812825715746, "tokens_seen": 646381568 }, { "epoch": 0.2, "learning_rate": 0.00040613841524573727, "loss": 2.6813, "theoretical_loss": 3.8103414985564292, "tokens_seen": 646447104 }, { "epoch": 0.2, "learning_rate": 0.0004061283851554664, "loss": 2.951, "theoretical_loss": 3.8103017197035145, "tokens_seen": 646512640 }, { "epoch": 0.2, "learning_rate": 0.00040611835506519563, "loss": 2.8612, "theoretical_loss": 3.810261946011636, "tokens_seen": 646578176 }, { "epoch": 0.2, "learning_rate": 0.00040610832497492476, "loss": 2.9858, "theoretical_loss": 3.8102221774796012, "tokens_seen": 646643712 }, { "epoch": 0.2, "learning_rate": 0.000406098294884654, "loss": 2.88, "theoretical_loss": 3.810182414106219, "tokens_seen": 646709248 }, { "epoch": 0.2, "learning_rate": 0.0004060882647943832, "loss": 2.7479, "theoretical_loss": 3.810142655890297, "tokens_seen": 646774784 }, { "epoch": 0.2, "learning_rate": 0.00040607823470411236, "loss": 3.0342, "theoretical_loss": 3.8101029028306437, "tokens_seen": 646840320 }, { "epoch": 0.2, "learning_rate": 0.00040606820461384154, "loss": 3.1329, "theoretical_loss": 3.810063154926068, "tokens_seen": 646905856 }, { "epoch": 0.2, "learning_rate": 0.0004060581745235707, "loss": 2.7757, "theoretical_loss": 3.8100234121753798, "tokens_seen": 646971392 }, { "epoch": 0.2, "learning_rate": 0.0004060481444332999, "loss": 2.9786, "theoretical_loss": 3.8099836745773885, "tokens_seen": 647036928 }, { "epoch": 0.2, "learning_rate": 0.00040603811434302914, "loss": 2.7087, "theoretical_loss": 3.8099439421309036, "tokens_seen": 647102464 }, { "epoch": 0.2, "objective/train/docs_used": 375146, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.229652166366577, "objective/train/theoretical_loss": 3.8099042148347366, "objective/train/tokens_used": 667628000, "theoretical_loss": 3.8099042148347366, "tokens_seen": 647168000 }, { "epoch": 0.2, "learning_rate": 0.00040602808425275826, "loss": 2.9167, "theoretical_loss": 3.8099042148347366, "tokens_seen": 647168000 }, { "epoch": 0.2, "learning_rate": 0.0004060180541624875, "loss": 2.4942, "theoretical_loss": 3.809864492687698, "tokens_seen": 647233536 }, { "epoch": 0.2, "learning_rate": 0.0004060080240722167, "loss": 2.9195, "theoretical_loss": 3.8098247756885995, "tokens_seen": 647299072 }, { "epoch": 0.2, "learning_rate": 0.00040599799398194586, "loss": 3.1232, "theoretical_loss": 3.8097850638362516, "tokens_seen": 647364608 }, { "epoch": 0.2, "learning_rate": 0.00040598796389167504, "loss": 2.9194, "theoretical_loss": 3.8097453571294686, "tokens_seen": 647430144 }, { "epoch": 0.2, "learning_rate": 0.0004059779338014042, "loss": 2.7781, "theoretical_loss": 3.8097056555670608, "tokens_seen": 647495680 }, { "epoch": 0.2, "learning_rate": 0.0004059679037111334, "loss": 2.9331, "theoretical_loss": 3.809665959147843, "tokens_seen": 647561216 }, { "epoch": 0.2, "learning_rate": 0.00040595787362086264, "loss": 3.0463, "theoretical_loss": 3.8096262678706267, "tokens_seen": 647626752 }, { "epoch": 0.2, "learning_rate": 0.00040594784353059177, "loss": 2.7725, "theoretical_loss": 3.8095865817342274, "tokens_seen": 647692288 }, { "epoch": 0.2, "learning_rate": 0.000405937813440321, "loss": 3.0406, "theoretical_loss": 3.809546900737457, "tokens_seen": 647757824 }, { "epoch": 0.2, "learning_rate": 0.00040592778335005013, "loss": 2.9272, "theoretical_loss": 3.809507224879132, "tokens_seen": 647823360 }, { "epoch": 0.2, "learning_rate": 0.00040591775325977936, "loss": 3.1577, "theoretical_loss": 3.809467554158067, "tokens_seen": 647888896 }, { "epoch": 0.2, "learning_rate": 0.00040590772316950854, "loss": 2.8766, "theoretical_loss": 3.809427888573077, "tokens_seen": 647954432 }, { "epoch": 0.2, "learning_rate": 0.0004058976930792377, "loss": 2.7179, "theoretical_loss": 3.8093882281229767, "tokens_seen": 648019968 }, { "epoch": 0.2, "learning_rate": 0.0004058876629889669, "loss": 2.9549, "theoretical_loss": 3.8093485728065835, "tokens_seen": 648085504 }, { "epoch": 0.2, "learning_rate": 0.0004058776328986961, "loss": 2.9438, "theoretical_loss": 3.809308922622713, "tokens_seen": 648151040 }, { "epoch": 0.2, "learning_rate": 0.00040586760280842527, "loss": 3.0333, "theoretical_loss": 3.8092692775701824, "tokens_seen": 648216576 }, { "epoch": 0.2, "learning_rate": 0.0004058575727181545, "loss": 2.7464, "theoretical_loss": 3.8092296376478094, "tokens_seen": 648282112 }, { "epoch": 0.2, "learning_rate": 0.00040584754262788363, "loss": 3.149, "theoretical_loss": 3.8091900028544106, "tokens_seen": 648347648 }, { "epoch": 0.2, "learning_rate": 0.00040583751253761287, "loss": 2.8448, "theoretical_loss": 3.809150373188805, "tokens_seen": 648413184 }, { "epoch": 0.2, "learning_rate": 0.00040582748244734205, "loss": 2.9585, "theoretical_loss": 3.80911074864981, "tokens_seen": 648478720 }, { "epoch": 0.2, "learning_rate": 0.00040581745235707123, "loss": 2.8377, "theoretical_loss": 3.8090711292362456, "tokens_seen": 648544256 }, { "epoch": 0.2, "learning_rate": 0.0004058074222668004, "loss": 2.8158, "theoretical_loss": 3.80903151494693, "tokens_seen": 648609792 }, { "epoch": 0.2, "learning_rate": 0.0004057973921765296, "loss": 3.0696, "theoretical_loss": 3.808991905780683, "tokens_seen": 648675328 }, { "epoch": 0.2, "learning_rate": 0.00040578736208625877, "loss": 2.875, "theoretical_loss": 3.8089523017363254, "tokens_seen": 648740864 }, { "epoch": 0.2, "objective/train/docs_used": 375860, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.106677770614624, "objective/train/theoretical_loss": 3.8089127028126764, "objective/train/tokens_used": 669266400, "theoretical_loss": 3.8089127028126764, "tokens_seen": 648806400 }, { "epoch": 0.2, "learning_rate": 0.000405777331995988, "loss": 2.8354, "theoretical_loss": 3.8089127028126764, "tokens_seen": 648806400 }, { "epoch": 0.2, "learning_rate": 0.00040576730190571713, "loss": 2.6935, "theoretical_loss": 3.8088731090085575, "tokens_seen": 648871936 }, { "epoch": 0.2, "learning_rate": 0.00040575727181544637, "loss": 2.7503, "theoretical_loss": 3.8088335203227897, "tokens_seen": 648937472 }, { "epoch": 0.2, "learning_rate": 0.0004057472417251755, "loss": 2.9742, "theoretical_loss": 3.8087939367541943, "tokens_seen": 649003008 }, { "epoch": 0.2, "learning_rate": 0.00040573721163490473, "loss": 3.0332, "theoretical_loss": 3.8087543583015933, "tokens_seen": 649068544 }, { "epoch": 0.2, "learning_rate": 0.0004057271815446339, "loss": 3.0165, "theoretical_loss": 3.8087147849638097, "tokens_seen": 649134080 }, { "epoch": 0.2, "learning_rate": 0.0004057171514543631, "loss": 3.0003, "theoretical_loss": 3.808675216739665, "tokens_seen": 649199616 }, { "epoch": 0.2, "learning_rate": 0.0004057071213640923, "loss": 3.0969, "theoretical_loss": 3.808635653627983, "tokens_seen": 649265152 }, { "epoch": 0.2, "learning_rate": 0.00040569709127382146, "loss": 2.8748, "theoretical_loss": 3.8085960956275873, "tokens_seen": 649330688 }, { "epoch": 0.2, "learning_rate": 0.00040568706118355064, "loss": 2.8892, "theoretical_loss": 3.8085565427373016, "tokens_seen": 649396224 }, { "epoch": 0.2, "learning_rate": 0.0004056770310932799, "loss": 2.6728, "theoretical_loss": 3.8085169949559505, "tokens_seen": 649461760 }, { "epoch": 0.2, "learning_rate": 0.000405667001003009, "loss": 2.8494, "theoretical_loss": 3.808477452282358, "tokens_seen": 649527296 }, { "epoch": 0.2, "learning_rate": 0.00040565697091273824, "loss": 3.0504, "theoretical_loss": 3.8084379147153493, "tokens_seen": 649592832 }, { "epoch": 0.2, "learning_rate": 0.0004056469408224674, "loss": 2.8025, "theoretical_loss": 3.808398382253751, "tokens_seen": 649658368 }, { "epoch": 0.2, "learning_rate": 0.0004056369107321966, "loss": 3.1074, "theoretical_loss": 3.8083588548963867, "tokens_seen": 649723904 }, { "epoch": 0.2, "learning_rate": 0.0004056268806419258, "loss": 2.8261, "theoretical_loss": 3.8083193326420846, "tokens_seen": 649789440 }, { "epoch": 0.2, "learning_rate": 0.00040561685055165496, "loss": 2.7663, "theoretical_loss": 3.808279815489671, "tokens_seen": 649854976 }, { "epoch": 0.2, "learning_rate": 0.00040560682046138414, "loss": 2.811, "theoretical_loss": 3.8082403034379713, "tokens_seen": 649920512 }, { "epoch": 0.2, "learning_rate": 0.0004055967903711134, "loss": 2.9304, "theoretical_loss": 3.808200796485815, "tokens_seen": 649986048 }, { "epoch": 0.2, "learning_rate": 0.0004055867602808425, "loss": 3.0942, "theoretical_loss": 3.8081612946320287, "tokens_seen": 650051584 }, { "epoch": 0.2, "learning_rate": 0.00040557673019057174, "loss": 2.9475, "theoretical_loss": 3.8081217978754407, "tokens_seen": 650117120 }, { "epoch": 0.2, "learning_rate": 0.00040556670010030087, "loss": 2.9979, "theoretical_loss": 3.80808230621488, "tokens_seen": 650182656 }, { "epoch": 0.2, "learning_rate": 0.0004055566700100301, "loss": 2.8176, "theoretical_loss": 3.8080428196491747, "tokens_seen": 650248192 }, { "epoch": 0.2, "learning_rate": 0.0004055466399197593, "loss": 2.9571, "theoretical_loss": 3.808003338177154, "tokens_seen": 650313728 }, { "epoch": 0.2, "learning_rate": 0.00040553660982948846, "loss": 2.8646, "theoretical_loss": 3.8079638617976492, "tokens_seen": 650379264 }, { "epoch": 0.2, "objective/train/docs_used": 377374, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.111595392227173, "objective/train/theoretical_loss": 3.8079243905094886, "objective/train/tokens_used": 670904800, "theoretical_loss": 3.8079243905094886, "tokens_seen": 650444800 }, { "epoch": 0.2, "learning_rate": 0.00040552657973921764, "loss": 2.8842, "theoretical_loss": 3.8079243905094886, "tokens_seen": 650444800 }, { "epoch": 0.2, "learning_rate": 0.0004055165496489469, "loss": 2.8977, "theoretical_loss": 3.8078849243115043, "tokens_seen": 650510336 }, { "epoch": 0.2, "learning_rate": 0.000405506519558676, "loss": 2.8774, "theoretical_loss": 3.8078454632025256, "tokens_seen": 650575872 }, { "epoch": 0.2, "learning_rate": 0.00040549648946840524, "loss": 2.9541, "theoretical_loss": 3.8078060071813846, "tokens_seen": 650641408 }, { "epoch": 0.2, "learning_rate": 0.00040548645937813437, "loss": 2.9887, "theoretical_loss": 3.8077665562469134, "tokens_seen": 650706944 }, { "epoch": 0.2, "learning_rate": 0.0004054764292878636, "loss": 2.8827, "theoretical_loss": 3.807727110397942, "tokens_seen": 650772480 }, { "epoch": 0.2, "learning_rate": 0.0004054663991975928, "loss": 2.8771, "theoretical_loss": 3.8076876696333057, "tokens_seen": 650838016 }, { "epoch": 0.2, "learning_rate": 0.00040545636910732197, "loss": 3.034, "theoretical_loss": 3.807648233951835, "tokens_seen": 650903552 }, { "epoch": 0.2, "learning_rate": 0.00040544633901705115, "loss": 3.0342, "theoretical_loss": 3.8076088033523643, "tokens_seen": 650969088 }, { "epoch": 0.2, "learning_rate": 0.00040543630892678033, "loss": 2.779, "theoretical_loss": 3.807569377833726, "tokens_seen": 651034624 }, { "epoch": 0.2, "learning_rate": 0.0004054262788365095, "loss": 3.1769, "theoretical_loss": 3.8075299573947556, "tokens_seen": 651100160 }, { "epoch": 0.2, "learning_rate": 0.00040541624874623874, "loss": 2.9482, "theoretical_loss": 3.8074905420342864, "tokens_seen": 651165696 }, { "epoch": 0.2, "learning_rate": 0.00040540621865596787, "loss": 2.8476, "theoretical_loss": 3.8074511317511535, "tokens_seen": 651231232 }, { "epoch": 0.2, "learning_rate": 0.0004053961885656971, "loss": 2.8999, "theoretical_loss": 3.807411726544191, "tokens_seen": 651296768 }, { "epoch": 0.2, "learning_rate": 0.0004053861584754263, "loss": 3.0428, "theoretical_loss": 3.8073723264122363, "tokens_seen": 651362304 }, { "epoch": 0.2, "learning_rate": 0.00040537612838515547, "loss": 2.8553, "theoretical_loss": 3.8073329313541233, "tokens_seen": 651427840 }, { "epoch": 0.2, "learning_rate": 0.0004053660982948847, "loss": 2.792, "theoretical_loss": 3.80729354136869, "tokens_seen": 651493376 }, { "epoch": 0.2, "learning_rate": 0.00040535606820461383, "loss": 2.9315, "theoretical_loss": 3.807254156454772, "tokens_seen": 651558912 }, { "epoch": 0.2, "learning_rate": 0.00040534603811434307, "loss": 3.0491, "theoretical_loss": 3.807214776611206, "tokens_seen": 651624448 }, { "epoch": 0.2, "learning_rate": 0.00040533600802407225, "loss": 3.1541, "theoretical_loss": 3.8071754018368305, "tokens_seen": 651689984 }, { "epoch": 0.2, "learning_rate": 0.00040532597793380143, "loss": 3.0628, "theoretical_loss": 3.807136032130483, "tokens_seen": 651755520 }, { "epoch": 0.2, "learning_rate": 0.0004053159478435306, "loss": 2.8765, "theoretical_loss": 3.8070966674910007, "tokens_seen": 651821056 }, { "epoch": 0.2, "learning_rate": 0.0004053059177532598, "loss": 2.9814, "theoretical_loss": 3.8070573079172227, "tokens_seen": 651886592 }, { "epoch": 0.2, "learning_rate": 0.00040529588766298897, "loss": 2.9982, "theoretical_loss": 3.807017953407989, "tokens_seen": 651952128 }, { "epoch": 0.2, "learning_rate": 0.0004052858575727182, "loss": 2.793, "theoretical_loss": 3.806978603962137, "tokens_seen": 652017664 }, { "epoch": 0.2, "objective/train/docs_used": 377873, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8976755142211914, "objective/train/theoretical_loss": 3.8069392595785083, "objective/train/tokens_used": 672543200, "theoretical_loss": 3.8069392595785083, "tokens_seen": 652083200 }, { "epoch": 0.2, "learning_rate": 0.00040527582748244733, "loss": 2.9065, "theoretical_loss": 3.8069392595785083, "tokens_seen": 652083200 }, { "epoch": 0.2, "learning_rate": 0.00040526579739217657, "loss": 2.8254, "theoretical_loss": 3.806899920255942, "tokens_seen": 652148736 }, { "epoch": 0.2, "learning_rate": 0.0004052557673019057, "loss": 2.8496, "theoretical_loss": 3.8068605859932783, "tokens_seen": 652214272 }, { "epoch": 0.2, "learning_rate": 0.00040524573721163493, "loss": 3.0224, "theoretical_loss": 3.806821256789359, "tokens_seen": 652279808 }, { "epoch": 0.2, "learning_rate": 0.0004052357071213641, "loss": 2.8427, "theoretical_loss": 3.806781932643024, "tokens_seen": 652345344 }, { "epoch": 0.2, "learning_rate": 0.0004052256770310933, "loss": 2.9331, "theoretical_loss": 3.8067426135531157, "tokens_seen": 652410880 }, { "epoch": 0.2, "learning_rate": 0.0004052156469408225, "loss": 2.758, "theoretical_loss": 3.8067032995184764, "tokens_seen": 652476416 }, { "epoch": 0.2, "learning_rate": 0.00040520561685055166, "loss": 3.0669, "theoretical_loss": 3.8066639905379485, "tokens_seen": 652541952 }, { "epoch": 0.2, "learning_rate": 0.00040519558676028084, "loss": 3.0724, "theoretical_loss": 3.806624686610374, "tokens_seen": 652607488 }, { "epoch": 0.2, "learning_rate": 0.0004051855566700101, "loss": 2.8816, "theoretical_loss": 3.8065853877345965, "tokens_seen": 652673024 }, { "epoch": 0.2, "learning_rate": 0.0004051755265797392, "loss": 2.8926, "theoretical_loss": 3.8065460939094597, "tokens_seen": 652738560 }, { "epoch": 0.2, "learning_rate": 0.00040516549648946844, "loss": 2.8759, "theoretical_loss": 3.8065068051338073, "tokens_seen": 652804096 }, { "epoch": 0.2, "learning_rate": 0.0004051554663991976, "loss": 2.7865, "theoretical_loss": 3.806467521406484, "tokens_seen": 652869632 }, { "epoch": 0.2, "learning_rate": 0.0004051454363089268, "loss": 2.9804, "theoretical_loss": 3.806428242726333, "tokens_seen": 652935168 }, { "epoch": 0.2, "learning_rate": 0.000405135406218656, "loss": 3.0628, "theoretical_loss": 3.8063889690922013, "tokens_seen": 653000704 }, { "epoch": 0.2, "learning_rate": 0.00040512537612838516, "loss": 2.6469, "theoretical_loss": 3.806349700502933, "tokens_seen": 653066240 }, { "epoch": 0.2, "learning_rate": 0.00040511534603811434, "loss": 2.8764, "theoretical_loss": 3.806310436957374, "tokens_seen": 653131776 }, { "epoch": 0.2, "learning_rate": 0.0004051053159478436, "loss": 2.9757, "theoretical_loss": 3.8062711784543715, "tokens_seen": 653197312 }, { "epoch": 0.2, "learning_rate": 0.0004050952858575727, "loss": 2.5247, "theoretical_loss": 3.8062319249927707, "tokens_seen": 653262848 }, { "epoch": 0.2, "learning_rate": 0.00040508525576730194, "loss": 2.9353, "theoretical_loss": 3.8061926765714196, "tokens_seen": 653328384 }, { "epoch": 0.2, "learning_rate": 0.00040507522567703107, "loss": 2.9652, "theoretical_loss": 3.806153433189165, "tokens_seen": 653393920 }, { "epoch": 0.2, "learning_rate": 0.0004050651955867603, "loss": 3.0484, "theoretical_loss": 3.806114194844855, "tokens_seen": 653459456 }, { "epoch": 0.2, "learning_rate": 0.0004050551654964895, "loss": 2.9003, "theoretical_loss": 3.8060749615373375, "tokens_seen": 653524992 }, { "epoch": 0.2, "learning_rate": 0.00040504513540621866, "loss": 2.933, "theoretical_loss": 3.8060357332654604, "tokens_seen": 653590528 }, { "epoch": 0.2, "learning_rate": 0.00040503510531594784, "loss": 2.7105, "theoretical_loss": 3.8059965100280735, "tokens_seen": 653656064 }, { "epoch": 0.2, "objective/train/docs_used": 379015, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.882786989212036, "objective/train/theoretical_loss": 3.8059572918240248, "objective/train/tokens_used": 674181600, "theoretical_loss": 3.8059572918240248, "tokens_seen": 653721600 }, { "epoch": 0.2, "learning_rate": 0.0004050250752256771, "loss": 2.7231, "theoretical_loss": 3.8059572918240248, "tokens_seen": 653721600 }, { "epoch": 0.2, "learning_rate": 0.0004050150451354062, "loss": 2.8368, "theoretical_loss": 3.8059180786521654, "tokens_seen": 653787136 }, { "epoch": 0.2, "learning_rate": 0.00040500501504513544, "loss": 2.9828, "theoretical_loss": 3.805878870511344, "tokens_seen": 653852672 }, { "epoch": 0.2, "learning_rate": 0.00040499498495486457, "loss": 2.8881, "theoretical_loss": 3.805839667400412, "tokens_seen": 653918208 }, { "epoch": 0.2, "learning_rate": 0.0004049849548645938, "loss": 2.8925, "theoretical_loss": 3.8058004693182195, "tokens_seen": 653983744 }, { "epoch": 0.2, "learning_rate": 0.000404974924774323, "loss": 2.9944, "theoretical_loss": 3.8057612762636173, "tokens_seen": 654049280 }, { "epoch": 0.2, "learning_rate": 0.00040496489468405217, "loss": 2.7396, "theoretical_loss": 3.8057220882354574, "tokens_seen": 654114816 }, { "epoch": 0.2, "learning_rate": 0.00040495486459378135, "loss": 2.8268, "theoretical_loss": 3.8056829052325916, "tokens_seen": 654180352 }, { "epoch": 0.2, "learning_rate": 0.00040494483450351053, "loss": 2.9557, "theoretical_loss": 3.805643727253872, "tokens_seen": 654245888 }, { "epoch": 0.2, "learning_rate": 0.0004049348044132397, "loss": 2.5606, "theoretical_loss": 3.8056045542981516, "tokens_seen": 654311424 }, { "epoch": 0.2, "learning_rate": 0.00040492477432296894, "loss": 2.9476, "theoretical_loss": 3.805565386364283, "tokens_seen": 654376960 }, { "epoch": 0.2, "learning_rate": 0.00040491474423269807, "loss": 2.8194, "theoretical_loss": 3.8055262234511193, "tokens_seen": 654442496 }, { "epoch": 0.2, "learning_rate": 0.0004049047141424273, "loss": 2.7053, "theoretical_loss": 3.8054870655575153, "tokens_seen": 654508032 }, { "epoch": 0.2, "learning_rate": 0.00040489468405215643, "loss": 2.7497, "theoretical_loss": 3.805447912682324, "tokens_seen": 654573568 }, { "epoch": 0.2, "learning_rate": 0.00040488465396188567, "loss": 2.5947, "theoretical_loss": 3.8054087648244, "tokens_seen": 654639104 }, { "epoch": 0.2, "learning_rate": 0.00040487462387161485, "loss": 3.0872, "theoretical_loss": 3.8053696219825994, "tokens_seen": 654704640 }, { "epoch": 0.2, "learning_rate": 0.00040486459378134403, "loss": 2.738, "theoretical_loss": 3.805330484155776, "tokens_seen": 654770176 }, { "epoch": 0.2, "learning_rate": 0.0004048545636910732, "loss": 2.5799, "theoretical_loss": 3.8052913513427864, "tokens_seen": 654835712 }, { "epoch": 0.2, "learning_rate": 0.00040484453360080245, "loss": 3.0017, "theoretical_loss": 3.8052522235424853, "tokens_seen": 654901248 }, { "epoch": 0.2, "learning_rate": 0.0004048345035105316, "loss": 2.7451, "theoretical_loss": 3.805213100753731, "tokens_seen": 654966784 }, { "epoch": 0.2, "learning_rate": 0.0004048244734202608, "loss": 2.7297, "theoretical_loss": 3.8051739829753792, "tokens_seen": 655032320 }, { "epoch": 0.2, "learning_rate": 0.00040481444332998994, "loss": 2.9863, "theoretical_loss": 3.8051348702062864, "tokens_seen": 655097856 }, { "epoch": 0.2, "learning_rate": 0.00040480441323971917, "loss": 2.7582, "theoretical_loss": 3.805095762445311, "tokens_seen": 655163392 }, { "epoch": 0.2, "learning_rate": 0.00040479438314944835, "loss": 3.0981, "theoretical_loss": 3.805056659691311, "tokens_seen": 655228928 }, { "epoch": 0.2, "learning_rate": 0.00040478435305917753, "loss": 2.759, "theoretical_loss": 3.805017561943144, "tokens_seen": 655294464 }, { "debugging/Self-BLEU-5": 0.36561994328618, "debugging/distinct-1-grams": 0.7722597331948876, "debugging/distinct-2-grams": 0.8492548173712738, "debugging/entropy-1-grams": 5.826808035476899, "debugging/entropy-2-grams": 6.590880961095115, "debugging/length": 495.27272727272725, "debugging/num_segments": 11, "epoch": 0.2, "objective/train/docs_used": 379813, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7408783435821533, "objective/train/theoretical_loss": 3.804978469199669, "objective/train/tokens_used": 675820000, "theoretical_loss": 3.804978469199669, "tokens_seen": 655360000 }, { "epoch": 0.2, "learning_rate": 0.0004047743229689067, "loss": 2.8606, "theoretical_loss": 3.804978469199669, "tokens_seen": 655360000 }, { "epoch": 0.2, "learning_rate": 0.0004047642928786359, "loss": 2.9838, "theoretical_loss": 3.804939381459745, "tokens_seen": 655425536 }, { "epoch": 0.2, "learning_rate": 0.0004047542627883651, "loss": 2.8588, "theoretical_loss": 3.8049002987222313, "tokens_seen": 655491072 }, { "epoch": 0.2, "learning_rate": 0.0004047442326980943, "loss": 2.91, "theoretical_loss": 3.8048612209859876, "tokens_seen": 655556608 }, { "epoch": 0.2, "learning_rate": 0.00040473420260782344, "loss": 2.7598, "theoretical_loss": 3.8048221482498743, "tokens_seen": 655622144 }, { "epoch": 0.2, "learning_rate": 0.0004047241725175527, "loss": 2.7256, "theoretical_loss": 3.8047830805127516, "tokens_seen": 655687680 }, { "epoch": 0.2, "learning_rate": 0.0004047141424272818, "loss": 2.8889, "theoretical_loss": 3.8047440177734804, "tokens_seen": 655753216 }, { "epoch": 0.2, "learning_rate": 0.00040470411233701104, "loss": 2.9605, "theoretical_loss": 3.8047049600309224, "tokens_seen": 655818752 }, { "epoch": 0.2, "learning_rate": 0.0004046940822467402, "loss": 3.0472, "theoretical_loss": 3.804665907283938, "tokens_seen": 655884288 }, { "epoch": 0.2, "learning_rate": 0.0004046840521564694, "loss": 2.8666, "theoretical_loss": 3.804626859531391, "tokens_seen": 655949824 }, { "epoch": 0.2, "learning_rate": 0.0004046740220661986, "loss": 2.9352, "theoretical_loss": 3.804587816772142, "tokens_seen": 656015360 }, { "epoch": 0.2, "learning_rate": 0.0004046639919759278, "loss": 2.6207, "theoretical_loss": 3.8045487790050547, "tokens_seen": 656080896 }, { "epoch": 0.2, "learning_rate": 0.000404653961885657, "loss": 2.9404, "theoretical_loss": 3.8045097462289923, "tokens_seen": 656146432 }, { "epoch": 0.2, "learning_rate": 0.0004046439317953862, "loss": 2.7605, "theoretical_loss": 3.804470718442818, "tokens_seen": 656211968 }, { "epoch": 0.2, "learning_rate": 0.00040463390170511536, "loss": 3.0043, "theoretical_loss": 3.8044316956453956, "tokens_seen": 656277504 }, { "epoch": 0.2, "learning_rate": 0.00040462387161484454, "loss": 2.8137, "theoretical_loss": 3.8043926778355894, "tokens_seen": 656343040 }, { "epoch": 0.2, "learning_rate": 0.0004046138415245738, "loss": 2.9791, "theoretical_loss": 3.804353665012264, "tokens_seen": 656408576 }, { "epoch": 0.2, "learning_rate": 0.0004046038114343029, "loss": 3.004, "theoretical_loss": 3.8043146571742845, "tokens_seen": 656474112 }, { "epoch": 0.2, "learning_rate": 0.00040459378134403214, "loss": 2.7808, "theoretical_loss": 3.8042756543205156, "tokens_seen": 656539648 }, { "epoch": 0.2, "learning_rate": 0.00040458375125376127, "loss": 3.0046, "theoretical_loss": 3.8042366564498242, "tokens_seen": 656605184 }, { "epoch": 0.2, "learning_rate": 0.0004045737211634905, "loss": 2.829, "theoretical_loss": 3.8041976635610752, "tokens_seen": 656670720 }, { "epoch": 0.2, "learning_rate": 0.0004045636910732197, "loss": 3.0518, "theoretical_loss": 3.8041586756531363, "tokens_seen": 656736256 }, { "epoch": 0.2, "learning_rate": 0.00040455366098294886, "loss": 3.0584, "theoretical_loss": 3.804119692724873, "tokens_seen": 656801792 }, { "epoch": 0.2, "learning_rate": 0.00040454363089267804, "loss": 3.0102, "theoretical_loss": 3.804080714775153, "tokens_seen": 656867328 }, { "epoch": 0.2, "learning_rate": 0.0004045336008024073, "loss": 2.9845, "theoretical_loss": 3.8040417418028443, "tokens_seen": 656932864 }, { "epoch": 0.2, "objective/train/docs_used": 381022, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.044368267059326, "objective/train/theoretical_loss": 3.8040027738068143, "objective/train/tokens_used": 677458400, "theoretical_loss": 3.8040027738068143, "tokens_seen": 656998400 }, { "epoch": 0.2, "learning_rate": 0.0004045235707121364, "loss": 2.9637, "theoretical_loss": 3.8040027738068143, "tokens_seen": 656998400 }, { "epoch": 0.2, "learning_rate": 0.00040451354062186564, "loss": 2.8752, "theoretical_loss": 3.8039638107859317, "tokens_seen": 657063936 }, { "epoch": 0.2, "learning_rate": 0.00040450351053159477, "loss": 3.2028, "theoretical_loss": 3.803924852739065, "tokens_seen": 657129472 }, { "epoch": 0.2, "learning_rate": 0.000404493480441324, "loss": 2.9586, "theoretical_loss": 3.803885899665083, "tokens_seen": 657195008 }, { "epoch": 0.2, "learning_rate": 0.0004044834503510532, "loss": 3.049, "theoretical_loss": 3.8038469515628557, "tokens_seen": 657260544 }, { "epoch": 0.2, "learning_rate": 0.00040447342026078237, "loss": 2.7938, "theoretical_loss": 3.803808008431252, "tokens_seen": 657326080 }, { "epoch": 0.2, "learning_rate": 0.00040446339017051155, "loss": 2.7174, "theoretical_loss": 3.8037690702691433, "tokens_seen": 657391616 }, { "epoch": 0.2, "learning_rate": 0.00040445336008024073, "loss": 2.754, "theoretical_loss": 3.8037301370753993, "tokens_seen": 657457152 }, { "epoch": 0.2, "learning_rate": 0.0004044433299899699, "loss": 3.0205, "theoretical_loss": 3.803691208848891, "tokens_seen": 657522688 }, { "epoch": 0.2, "learning_rate": 0.00040443329989969915, "loss": 2.9776, "theoretical_loss": 3.8036522855884893, "tokens_seen": 657588224 }, { "epoch": 0.2, "learning_rate": 0.00040442326980942827, "loss": 2.5956, "theoretical_loss": 3.803613367293067, "tokens_seen": 657653760 }, { "epoch": 0.2, "learning_rate": 0.0004044132397191575, "loss": 2.9704, "theoretical_loss": 3.803574453961495, "tokens_seen": 657719296 }, { "epoch": 0.2, "learning_rate": 0.00040440320962888663, "loss": 3.0798, "theoretical_loss": 3.803535545592646, "tokens_seen": 657784832 }, { "epoch": 0.2, "learning_rate": 0.00040439317953861587, "loss": 2.8397, "theoretical_loss": 3.803496642185393, "tokens_seen": 657850368 }, { "epoch": 0.2, "learning_rate": 0.00040438314944834505, "loss": 2.7924, "theoretical_loss": 3.8034577437386083, "tokens_seen": 657915904 }, { "epoch": 0.2, "learning_rate": 0.00040437311935807423, "loss": 2.82, "theoretical_loss": 3.8034188502511665, "tokens_seen": 657981440 }, { "epoch": 0.2, "learning_rate": 0.0004043630892678034, "loss": 2.6594, "theoretical_loss": 3.803379961721941, "tokens_seen": 658046976 }, { "epoch": 0.2, "learning_rate": 0.00040435305917753265, "loss": 2.835, "theoretical_loss": 3.8033410781498054, "tokens_seen": 658112512 }, { "epoch": 0.2, "learning_rate": 0.0004043430290872618, "loss": 2.9879, "theoretical_loss": 3.803302199533635, "tokens_seen": 658178048 }, { "epoch": 0.2, "learning_rate": 0.000404332998996991, "loss": 3.0102, "theoretical_loss": 3.8032633258723045, "tokens_seen": 658243584 }, { "epoch": 0.2, "learning_rate": 0.00040432296890672014, "loss": 2.7771, "theoretical_loss": 3.8032244571646894, "tokens_seen": 658309120 }, { "epoch": 0.2, "learning_rate": 0.00040431293881644937, "loss": 2.8471, "theoretical_loss": 3.8031855934096654, "tokens_seen": 658374656 }, { "epoch": 0.2, "learning_rate": 0.00040430290872617855, "loss": 2.8971, "theoretical_loss": 3.8031467346061083, "tokens_seen": 658440192 }, { "epoch": 0.2, "learning_rate": 0.00040429287863590773, "loss": 2.9073, "theoretical_loss": 3.8031078807528944, "tokens_seen": 658505728 }, { "epoch": 0.2, "learning_rate": 0.0004042828485456369, "loss": 2.7519, "theoretical_loss": 3.803069031848901, "tokens_seen": 658571264 }, { "epoch": 0.2, "objective/train/docs_used": 381585, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.120889663696289, "objective/train/theoretical_loss": 3.803030187893005, "objective/train/tokens_used": 679096800, "theoretical_loss": 3.803030187893005, "tokens_seen": 658636800 }, { "epoch": 0.2, "learning_rate": 0.0004042728184553661, "loss": 2.7675, "theoretical_loss": 3.803030187893005, "tokens_seen": 658636800 }, { "epoch": 0.2, "learning_rate": 0.0004042627883650953, "loss": 2.805, "theoretical_loss": 3.802991348884084, "tokens_seen": 658702336 }, { "epoch": 0.2, "learning_rate": 0.0004042527582748245, "loss": 2.8179, "theoretical_loss": 3.8029525148210146, "tokens_seen": 658767872 }, { "epoch": 0.2, "learning_rate": 0.00040424272818455364, "loss": 3.0148, "theoretical_loss": 3.802913685702677, "tokens_seen": 658833408 }, { "epoch": 0.2, "learning_rate": 0.0004042326980942829, "loss": 2.8772, "theoretical_loss": 3.8028748615279495, "tokens_seen": 658898944 }, { "epoch": 0.2, "learning_rate": 0.000404222668004012, "loss": 3.0401, "theoretical_loss": 3.80283604229571, "tokens_seen": 658964480 }, { "epoch": 0.2, "learning_rate": 0.00040421263791374124, "loss": 2.5957, "theoretical_loss": 3.802797228004839, "tokens_seen": 659030016 }, { "epoch": 0.2, "learning_rate": 0.0004042026078234704, "loss": 2.8999, "theoretical_loss": 3.802758418654215, "tokens_seen": 659095552 }, { "epoch": 0.2, "learning_rate": 0.0004041925777331996, "loss": 2.931, "theoretical_loss": 3.8027196142427195, "tokens_seen": 659161088 }, { "epoch": 0.2, "learning_rate": 0.0004041825476429288, "loss": 2.8085, "theoretical_loss": 3.802680814769232, "tokens_seen": 659226624 }, { "epoch": 0.2, "learning_rate": 0.000404172517552658, "loss": 2.7624, "theoretical_loss": 3.802642020232633, "tokens_seen": 659292160 }, { "epoch": 0.2, "learning_rate": 0.00040416248746238714, "loss": 2.8777, "theoretical_loss": 3.8026032306318047, "tokens_seen": 659357696 }, { "epoch": 0.2, "learning_rate": 0.0004041524573721164, "loss": 3.0243, "theoretical_loss": 3.8025644459656283, "tokens_seen": 659423232 }, { "epoch": 0.2, "learning_rate": 0.0004041424272818455, "loss": 2.8105, "theoretical_loss": 3.802525666232985, "tokens_seen": 659488768 }, { "epoch": 0.2, "learning_rate": 0.00040413239719157474, "loss": 2.7355, "theoretical_loss": 3.802486891432758, "tokens_seen": 659554304 }, { "epoch": 0.2, "learning_rate": 0.0004041223671013039, "loss": 2.9355, "theoretical_loss": 3.802448121563829, "tokens_seen": 659619840 }, { "epoch": 0.2, "learning_rate": 0.0004041123370110331, "loss": 2.7167, "theoretical_loss": 3.802409356625082, "tokens_seen": 659685376 }, { "epoch": 0.2, "learning_rate": 0.0004041023069207623, "loss": 2.9943, "theoretical_loss": 3.8023705966153996, "tokens_seen": 659750912 }, { "epoch": 0.2, "learning_rate": 0.00040409227683049147, "loss": 2.7669, "theoretical_loss": 3.802331841533666, "tokens_seen": 659816448 }, { "epoch": 0.2, "learning_rate": 0.00040408224674022065, "loss": 2.9835, "theoretical_loss": 3.8022930913787647, "tokens_seen": 659881984 }, { "epoch": 0.2, "learning_rate": 0.0004040722166499499, "loss": 2.847, "theoretical_loss": 3.8022543461495806, "tokens_seen": 659947520 }, { "epoch": 0.2, "learning_rate": 0.000404062186559679, "loss": 2.982, "theoretical_loss": 3.8022156058449985, "tokens_seen": 660013056 }, { "epoch": 0.2, "learning_rate": 0.00040405215646940824, "loss": 2.9005, "theoretical_loss": 3.802176870463904, "tokens_seen": 660078592 }, { "epoch": 0.2, "learning_rate": 0.00040404212637913737, "loss": 2.7324, "theoretical_loss": 3.802138140005182, "tokens_seen": 660144128 }, { "epoch": 0.2, "learning_rate": 0.0004040320962888666, "loss": 2.912, "theoretical_loss": 3.8020994144677185, "tokens_seen": 660209664 }, { "epoch": 0.2, "objective/train/docs_used": 382875, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.6055595874786377, "objective/train/theoretical_loss": 3.8020606938504, "objective/train/tokens_used": 680735200, "theoretical_loss": 3.8020606938504, "tokens_seen": 660275200 }, { "epoch": 0.2, "learning_rate": 0.0004040220661985958, "loss": 2.9253, "theoretical_loss": 3.8020606938504, "tokens_seen": 660275200 }, { "epoch": 0.2, "learning_rate": 0.00040401203610832497, "loss": 3.18, "theoretical_loss": 3.8020219781521125, "tokens_seen": 660340736 }, { "epoch": 0.2, "learning_rate": 0.00040400200601805415, "loss": 3.0083, "theoretical_loss": 3.8019832673717437, "tokens_seen": 660406272 }, { "epoch": 0.2, "learning_rate": 0.0004039919759277834, "loss": 2.6783, "theoretical_loss": 3.801944561508181, "tokens_seen": 660471808 }, { "epoch": 0.2, "learning_rate": 0.0004039819458375125, "loss": 2.9787, "theoretical_loss": 3.801905860560311, "tokens_seen": 660537344 }, { "epoch": 0.2, "learning_rate": 0.00040397191574724175, "loss": 2.9549, "theoretical_loss": 3.8018671645270237, "tokens_seen": 660602880 }, { "epoch": 0.2, "learning_rate": 0.0004039618856569709, "loss": 2.8998, "theoretical_loss": 3.8018284734072054, "tokens_seen": 660668416 }, { "epoch": 0.2, "learning_rate": 0.0004039518555667001, "loss": 2.953, "theoretical_loss": 3.8017897871997466, "tokens_seen": 660733952 }, { "epoch": 0.2, "learning_rate": 0.0004039418254764293, "loss": 2.8189, "theoretical_loss": 3.801751105903536, "tokens_seen": 660799488 }, { "epoch": 0.2, "learning_rate": 0.00040393179538615847, "loss": 2.91, "theoretical_loss": 3.8017124295174622, "tokens_seen": 660865024 }, { "epoch": 0.2, "learning_rate": 0.00040392176529588765, "loss": 2.864, "theoretical_loss": 3.8016737580404163, "tokens_seen": 660930560 }, { "epoch": 0.2, "learning_rate": 0.00040391173520561683, "loss": 2.773, "theoretical_loss": 3.8016350914712884, "tokens_seen": 660996096 }, { "epoch": 0.2, "learning_rate": 0.00040390170511534607, "loss": 2.6822, "theoretical_loss": 3.801596429808968, "tokens_seen": 661061632 }, { "epoch": 0.2, "learning_rate": 0.00040389167502507525, "loss": 2.7872, "theoretical_loss": 3.8015577730523473, "tokens_seen": 661127168 }, { "epoch": 0.2, "learning_rate": 0.00040388164493480443, "loss": 2.9026, "theoretical_loss": 3.801519121200317, "tokens_seen": 661192704 }, { "epoch": 0.2, "learning_rate": 0.0004038716148445336, "loss": 2.9067, "theoretical_loss": 3.801480474251769, "tokens_seen": 661258240 }, { "epoch": 0.2, "learning_rate": 0.00040386158475426285, "loss": 2.8634, "theoretical_loss": 3.801441832205595, "tokens_seen": 661323776 }, { "epoch": 0.2, "learning_rate": 0.000403851554663992, "loss": 2.8866, "theoretical_loss": 3.801403195060688, "tokens_seen": 661389312 }, { "epoch": 0.2, "learning_rate": 0.0004038415245737212, "loss": 2.7794, "theoretical_loss": 3.801364562815941, "tokens_seen": 661454848 }, { "epoch": 0.2, "learning_rate": 0.00040383149448345034, "loss": 2.8431, "theoretical_loss": 3.8013259354702456, "tokens_seen": 661520384 }, { "epoch": 0.2, "learning_rate": 0.00040382146439317957, "loss": 2.9924, "theoretical_loss": 3.801287313022497, "tokens_seen": 661585920 }, { "epoch": 0.2, "learning_rate": 0.00040381143430290875, "loss": 2.7929, "theoretical_loss": 3.801248695471588, "tokens_seen": 661651456 }, { "epoch": 0.2, "learning_rate": 0.00040380140421263794, "loss": 3.0859, "theoretical_loss": 3.8012100828164135, "tokens_seen": 661716992 }, { "epoch": 0.2, "learning_rate": 0.0004037913741223671, "loss": 2.7847, "theoretical_loss": 3.801171475055867, "tokens_seen": 661782528 }, { "epoch": 0.2, "learning_rate": 0.0004037813440320963, "loss": 2.8057, "theoretical_loss": 3.8011328721888447, "tokens_seen": 661848064 }, { "epoch": 0.2, "objective/train/docs_used": 383560, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.1271653175354004, "objective/train/theoretical_loss": 3.8010942742142415, "objective/train/tokens_used": 682373600, "theoretical_loss": 3.8010942742142415, "tokens_seen": 661913600 }, { "epoch": 0.2, "learning_rate": 0.0004037713139418255, "loss": 2.7091, "theoretical_loss": 3.8010942742142415, "tokens_seen": 661913600 }, { "epoch": 0.2, "learning_rate": 0.0004037612838515547, "loss": 2.8806, "theoretical_loss": 3.8010556811309524, "tokens_seen": 661979136 }, { "epoch": 0.2, "learning_rate": 0.00040375125376128384, "loss": 2.941, "theoretical_loss": 3.801017092937874, "tokens_seen": 662044672 }, { "epoch": 0.2, "learning_rate": 0.0004037412236710131, "loss": 2.9441, "theoretical_loss": 3.800978509633903, "tokens_seen": 662110208 }, { "epoch": 0.2, "learning_rate": 0.0004037311935807422, "loss": 2.7173, "theoretical_loss": 3.8009399312179353, "tokens_seen": 662175744 }, { "epoch": 0.2, "learning_rate": 0.00040372116349047144, "loss": 2.8, "theoretical_loss": 3.800901357688868, "tokens_seen": 662241280 }, { "epoch": 0.2, "learning_rate": 0.0004037111334002006, "loss": 2.9218, "theoretical_loss": 3.800862789045599, "tokens_seen": 662306816 }, { "epoch": 0.2, "learning_rate": 0.0004037011033099298, "loss": 2.7922, "theoretical_loss": 3.800824225287026, "tokens_seen": 662372352 }, { "epoch": 0.2, "learning_rate": 0.000403691073219659, "loss": 2.9749, "theoretical_loss": 3.8007856664120476, "tokens_seen": 662437888 }, { "epoch": 0.2, "learning_rate": 0.0004036810431293882, "loss": 2.8713, "theoretical_loss": 3.800747112419561, "tokens_seen": 662503424 }, { "epoch": 0.2, "learning_rate": 0.00040367101303911734, "loss": 2.9567, "theoretical_loss": 3.8007085633084663, "tokens_seen": 662568960 }, { "epoch": 0.2, "learning_rate": 0.0004036609829488466, "loss": 2.9524, "theoretical_loss": 3.8006700190776623, "tokens_seen": 662634496 }, { "epoch": 0.2, "learning_rate": 0.0004036509528585757, "loss": 2.9928, "theoretical_loss": 3.8006314797260488, "tokens_seen": 662700032 }, { "epoch": 0.2, "learning_rate": 0.00040364092276830494, "loss": 2.9063, "theoretical_loss": 3.800592945252525, "tokens_seen": 662765568 }, { "epoch": 0.2, "learning_rate": 0.0004036308926780341, "loss": 2.9209, "theoretical_loss": 3.8005544156559923, "tokens_seen": 662831104 }, { "epoch": 0.2, "learning_rate": 0.0004036208625877633, "loss": 2.8781, "theoretical_loss": 3.80051589093535, "tokens_seen": 662896640 }, { "epoch": 0.2, "learning_rate": 0.0004036108324974925, "loss": 2.7279, "theoretical_loss": 3.8004773710895003, "tokens_seen": 662962176 }, { "epoch": 0.2, "learning_rate": 0.00040360080240722167, "loss": 2.896, "theoretical_loss": 3.800438856117344, "tokens_seen": 663027712 }, { "epoch": 0.2, "learning_rate": 0.00040359077231695085, "loss": 2.6029, "theoretical_loss": 3.800400346017783, "tokens_seen": 663093248 }, { "epoch": 0.2, "learning_rate": 0.0004035807422266801, "loss": 2.9801, "theoretical_loss": 3.8003618407897193, "tokens_seen": 663158784 }, { "epoch": 0.2, "learning_rate": 0.0004035707121364092, "loss": 3.0723, "theoretical_loss": 3.800323340432055, "tokens_seen": 663224320 }, { "epoch": 0.2, "learning_rate": 0.00040356068204613844, "loss": 2.7635, "theoretical_loss": 3.8002848449436932, "tokens_seen": 663289856 }, { "epoch": 0.2, "learning_rate": 0.00040355065195586757, "loss": 2.8274, "theoretical_loss": 3.800246354323538, "tokens_seen": 663355392 }, { "epoch": 0.2, "learning_rate": 0.0004035406218655968, "loss": 2.9454, "theoretical_loss": 3.8002078685704914, "tokens_seen": 663420928 }, { "epoch": 0.2, "learning_rate": 0.000403530591775326, "loss": 2.9828, "theoretical_loss": 3.8001693876834572, "tokens_seen": 663486464 }, { "epoch": 0.2, "objective/train/docs_used": 384103, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.467216730117798, "objective/train/theoretical_loss": 3.8001309116613413, "objective/train/tokens_used": 684012000, "theoretical_loss": 3.8001309116613413, "tokens_seen": 663552000 }, { "epoch": 0.2, "learning_rate": 0.00040352056168505517, "loss": 2.9179, "theoretical_loss": 3.8001309116613413, "tokens_seen": 663552000 }, { "epoch": 0.2, "learning_rate": 0.00040351053159478435, "loss": 2.7601, "theoretical_loss": 3.8000924405030467, "tokens_seen": 663617536 }, { "epoch": 0.2, "learning_rate": 0.0004035005015045136, "loss": 2.8075, "theoretical_loss": 3.8000539742074784, "tokens_seen": 663683072 }, { "epoch": 0.2, "learning_rate": 0.0004034904714142427, "loss": 2.9031, "theoretical_loss": 3.8000155127735424, "tokens_seen": 663748608 }, { "epoch": 0.2, "learning_rate": 0.00040348044132397195, "loss": 2.7523, "theoretical_loss": 3.7999770562001443, "tokens_seen": 663814144 }, { "epoch": 0.2, "learning_rate": 0.0004034704112337011, "loss": 2.8653, "theoretical_loss": 3.799938604486189, "tokens_seen": 663879680 }, { "epoch": 0.2, "learning_rate": 0.0004034603811434303, "loss": 2.7593, "theoretical_loss": 3.799900157630584, "tokens_seen": 663945216 }, { "epoch": 0.2, "learning_rate": 0.0004034503510531595, "loss": 2.903, "theoretical_loss": 3.799861715632236, "tokens_seen": 664010752 }, { "epoch": 0.2, "learning_rate": 0.00040344032096288867, "loss": 3.0078, "theoretical_loss": 3.7998232784900514, "tokens_seen": 664076288 }, { "epoch": 0.2, "learning_rate": 0.00040343029087261785, "loss": 2.588, "theoretical_loss": 3.7997848462029378, "tokens_seen": 664141824 }, { "epoch": 0.2, "learning_rate": 0.00040342026078234703, "loss": 2.7286, "theoretical_loss": 3.7997464187698027, "tokens_seen": 664207360 }, { "epoch": 0.2, "learning_rate": 0.0004034102306920762, "loss": 2.8097, "theoretical_loss": 3.799707996189555, "tokens_seen": 664272896 }, { "epoch": 0.2, "learning_rate": 0.00040340020060180545, "loss": 2.8803, "theoretical_loss": 3.7996695784611023, "tokens_seen": 664338432 }, { "epoch": 0.2, "learning_rate": 0.0004033901705115346, "loss": 2.8301, "theoretical_loss": 3.799631165583354, "tokens_seen": 664403968 }, { "epoch": 0.2, "learning_rate": 0.0004033801404212638, "loss": 2.8001, "theoretical_loss": 3.7995927575552186, "tokens_seen": 664469504 }, { "epoch": 0.2, "learning_rate": 0.000403370110330993, "loss": 2.9368, "theoretical_loss": 3.7995543543756063, "tokens_seen": 664535040 } ], "max_steps": 50354, "num_train_epochs": 9223372036854775807, "total_flos": 3.3913615417344e+17, "trial_name": null, "trial_params": null }