{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.10626587637595258, "global_step": 251, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 4.1666666666666665e-05, "loss": 3.0643, "theoretical_loss": 3.321567680436603, "tokens_seen": 2990538752 }, { "epoch": 0.0, "learning_rate": 8.333333333333333e-05, "loss": 3.0798, "theoretical_loss": 3.3215564803546, "tokens_seen": 2990669824 }, { "epoch": 0.0, "learning_rate": 0.000125, "loss": 2.8688, "theoretical_loss": 3.321545280900887, "tokens_seen": 2990800896 }, { "epoch": 0.0, "learning_rate": 0.00016666666666666666, "loss": 2.7194, "theoretical_loss": 3.3215340820754022, "tokens_seen": 2990931968 }, { "epoch": 0.0, "learning_rate": 0.00020833333333333335, "loss": 2.6193, "theoretical_loss": 3.3215228838780817, "tokens_seen": 2991063040 }, { "epoch": 0.0, "learning_rate": 0.00025, "loss": 2.8571, "theoretical_loss": 3.3215116863088636, "tokens_seen": 2991194112 }, { "epoch": 0.0, "learning_rate": 0.0002916666666666667, "loss": 2.7571, "theoretical_loss": 3.3215004893676854, "tokens_seen": 2991325184 }, { "epoch": 0.0, "learning_rate": 0.0003333333333333333, "loss": 2.8877, "theoretical_loss": 3.321489293054483, "tokens_seen": 2991456256 }, { "epoch": 0.0, "learning_rate": 0.000375, "loss": 2.714, "theoretical_loss": 3.321478097369195, "tokens_seen": 2991587328 }, { "epoch": 0.0, "learning_rate": 0.0004166666666666667, "loss": 2.6564, "theoretical_loss": 3.321466902311758, "tokens_seen": 2991718400 }, { "epoch": 0.0, "learning_rate": 0.0004583333333333333, "loss": 2.5638, "theoretical_loss": 3.3214557078821096, "tokens_seen": 2991849472 }, { "epoch": 0.01, "learning_rate": 0.0005, "loss": 2.4354, "theoretical_loss": 3.321444514080187, "tokens_seen": 2991980544 }, { "epoch": 0.01, "objective/train/docs_used": 1640856, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5749809741973877, "objective/train/theoretical_loss": 3.321438917414603, "objective/train/tokens_used": 22097376, "theoretical_loss": 3.321438917414603, "tokens_seen": 2992046080 }, { "epoch": 0.01, "learning_rate": 0.0005416666666666666, "loss": 2.5713, "theoretical_loss": 3.321433320905927, "tokens_seen": 2992111616 }, { "epoch": 0.01, "learning_rate": 0.0005833333333333334, "loss": 2.4812, "theoretical_loss": 3.3214221283592678, "tokens_seen": 2992242688 }, { "epoch": 0.01, "learning_rate": 0.000625, "loss": 2.7622, "theoretical_loss": 3.321410936440146, "tokens_seen": 2992373760 }, { "epoch": 0.01, "learning_rate": 0.0006666666666666666, "loss": 2.609, "theoretical_loss": 3.3213997451485, "tokens_seen": 2992504832 }, { "epoch": 0.01, "learning_rate": 0.0007083333333333334, "loss": 2.4537, "theoretical_loss": 3.3213885544842654, "tokens_seen": 2992635904 }, { "epoch": 0.01, "learning_rate": 0.00075, "loss": 2.4831, "theoretical_loss": 3.321377364447381, "tokens_seen": 2992766976 }, { "epoch": 0.01, "learning_rate": 0.0007916666666666666, "loss": 2.5607, "theoretical_loss": 3.3213661750377836, "tokens_seen": 2992898048 }, { "epoch": 0.01, "learning_rate": 0.0008333333333333334, "loss": 2.5759, "theoretical_loss": 3.3213549862554106, "tokens_seen": 2993029120 }, { "epoch": 0.01, "learning_rate": 0.000875, "loss": 2.3106, "theoretical_loss": 3.3213437981001994, "tokens_seen": 2993160192 }, { "epoch": 0.01, "learning_rate": 0.0009166666666666666, "loss": 2.5471, "theoretical_loss": 3.3213326105720875, "tokens_seen": 2993291264 }, { "epoch": 0.01, "learning_rate": 0.0009583333333333334, "loss": 2.602, "theoretical_loss": 3.3213214236710122, "tokens_seen": 2993422336 }, { "epoch": 0.01, "learning_rate": 0.001, "loss": 2.5877, "theoretical_loss": 3.321310237396911, "tokens_seen": 2993553408 }, { "epoch": 0.01, "objective/train/docs_used": 1641461, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.557373523712158, "objective/train/theoretical_loss": 3.3212990517497207, "objective/train/tokens_used": 23735776, "theoretical_loss": 3.3212990517497207, "tokens_seen": 2993684480 }, { "epoch": 0.01, "learning_rate": 0.0009995722840034217, "loss": 2.7042, "theoretical_loss": 3.3212990517497207, "tokens_seen": 2993684480 }, { "epoch": 0.01, "learning_rate": 0.0009991445680068436, "loss": 2.5234, "theoretical_loss": 3.3212878667293797, "tokens_seen": 2993815552 }, { "epoch": 0.01, "learning_rate": 0.0009987168520102653, "loss": 2.5502, "theoretical_loss": 3.321276682335825, "tokens_seen": 2993946624 }, { "epoch": 0.01, "learning_rate": 0.000998289136013687, "loss": 2.6849, "theoretical_loss": 3.3212654985689936, "tokens_seen": 2994077696 }, { "epoch": 0.01, "learning_rate": 0.0009978614200171086, "loss": 2.6348, "theoretical_loss": 3.3212543154288237, "tokens_seen": 2994208768 }, { "epoch": 0.01, "learning_rate": 0.0009974337040205303, "loss": 2.6793, "theoretical_loss": 3.3212431329152525, "tokens_seen": 2994339840 }, { "epoch": 0.01, "learning_rate": 0.0009970059880239522, "loss": 2.6212, "theoretical_loss": 3.321231951028217, "tokens_seen": 2994470912 }, { "epoch": 0.01, "learning_rate": 0.0009965782720273739, "loss": 2.629, "theoretical_loss": 3.3212207697676552, "tokens_seen": 2994601984 }, { "epoch": 0.01, "learning_rate": 0.0009961505560307955, "loss": 2.5865, "theoretical_loss": 3.3212095891335043, "tokens_seen": 2994733056 }, { "epoch": 0.01, "learning_rate": 0.0009957228400342174, "loss": 2.667, "theoretical_loss": 3.321198409125702, "tokens_seen": 2994864128 }, { "epoch": 0.01, "learning_rate": 0.000995295124037639, "loss": 2.55, "theoretical_loss": 3.321187229744186, "tokens_seen": 2994995200 }, { "epoch": 0.02, "learning_rate": 0.0009948674080410608, "loss": 2.6324, "theoretical_loss": 3.321176050988893, "tokens_seen": 2995126272 }, { "epoch": 0.02, "learning_rate": 0.0009944396920444824, "loss": 2.8406, "theoretical_loss": 3.3211648728597614, "tokens_seen": 2995257344 }, { "epoch": 0.02, "objective/train/docs_used": 1642666, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8021185398101807, "objective/train/theoretical_loss": 3.3211592840299864, "objective/train/tokens_used": 25374176, "theoretical_loss": 3.3211592840299864, "tokens_seen": 2995322880 }, { "epoch": 0.02, "learning_rate": 0.0009940119760479041, "loss": 2.5438, "theoretical_loss": 3.3211536953567284, "tokens_seen": 2995388416 }, { "epoch": 0.02, "learning_rate": 0.000993584260051326, "loss": 2.7618, "theoretical_loss": 3.321142518479731, "tokens_seen": 2995519488 }, { "epoch": 0.02, "learning_rate": 0.0009931565440547477, "loss": 2.6638, "theoretical_loss": 3.321131342228708, "tokens_seen": 2995650560 }, { "epoch": 0.02, "learning_rate": 0.0009927288280581694, "loss": 2.7287, "theoretical_loss": 3.321120166603596, "tokens_seen": 2995781632 }, { "epoch": 0.02, "learning_rate": 0.000992301112061591, "loss": 2.5832, "theoretical_loss": 3.3211089916043326, "tokens_seen": 2995912704 }, { "epoch": 0.02, "learning_rate": 0.0009918733960650127, "loss": 2.6355, "theoretical_loss": 3.3210978172308554, "tokens_seen": 2996043776 }, { "epoch": 0.02, "learning_rate": 0.0009914456800684346, "loss": 2.5139, "theoretical_loss": 3.3210866434831026, "tokens_seen": 2996174848 }, { "epoch": 0.02, "learning_rate": 0.0009910179640718563, "loss": 2.5879, "theoretical_loss": 3.3210754703610106, "tokens_seen": 2996305920 }, { "epoch": 0.02, "learning_rate": 0.0009905902480752782, "loss": 2.5935, "theoretical_loss": 3.321064297864518, "tokens_seen": 2996436992 }, { "epoch": 0.02, "learning_rate": 0.0009901625320786998, "loss": 2.6176, "theoretical_loss": 3.3210531259935627, "tokens_seen": 2996568064 }, { "epoch": 0.02, "learning_rate": 0.0009897348160821215, "loss": 2.6405, "theoretical_loss": 3.321041954748081, "tokens_seen": 2996699136 }, { "epoch": 0.02, "learning_rate": 0.0009893071000855432, "loss": 2.7274, "theoretical_loss": 3.321030784128012, "tokens_seen": 2996830208 }, { "epoch": 0.02, "objective/train/docs_used": 1643300, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.023165464401245, "objective/train/theoretical_loss": 3.321019614133292, "objective/train/tokens_used": 27012576, "theoretical_loss": 3.321019614133292, "tokens_seen": 2996961280 }, { "epoch": 0.02, "learning_rate": 0.0009888793840889649, "loss": 2.7003, "theoretical_loss": 3.321019614133292, "tokens_seen": 2996961280 }, { "epoch": 0.02, "learning_rate": 0.0009884516680923865, "loss": 2.7436, "theoretical_loss": 3.3210084447638595, "tokens_seen": 2997092352 }, { "epoch": 0.02, "learning_rate": 0.0009880239520958084, "loss": 2.5873, "theoretical_loss": 3.320997276019652, "tokens_seen": 2997223424 }, { "epoch": 0.02, "learning_rate": 0.00098759623609923, "loss": 2.5974, "theoretical_loss": 3.3209861079006067, "tokens_seen": 2997354496 }, { "epoch": 0.02, "learning_rate": 0.000987168520102652, "loss": 2.5806, "theoretical_loss": 3.320974940406662, "tokens_seen": 2997485568 }, { "epoch": 0.02, "learning_rate": 0.0009867408041060737, "loss": 2.6771, "theoretical_loss": 3.320963773537755, "tokens_seen": 2997616640 }, { "epoch": 0.02, "learning_rate": 0.0009863130881094953, "loss": 2.7313, "theoretical_loss": 3.320952607293824, "tokens_seen": 2997747712 }, { "epoch": 0.02, "learning_rate": 0.000985885372112917, "loss": 2.7302, "theoretical_loss": 3.320941441674806, "tokens_seen": 2997878784 }, { "epoch": 0.02, "learning_rate": 0.0009854576561163387, "loss": 2.893, "theoretical_loss": 3.320930276680639, "tokens_seen": 2998009856 }, { "epoch": 0.02, "learning_rate": 0.0009850299401197606, "loss": 2.6886, "theoretical_loss": 3.3209191123112607, "tokens_seen": 2998140928 }, { "epoch": 0.03, "learning_rate": 0.0009846022241231823, "loss": 2.7801, "theoretical_loss": 3.320907948566609, "tokens_seen": 2998272000 }, { "epoch": 0.03, "learning_rate": 0.000984174508126604, "loss": 2.6538, "theoretical_loss": 3.3208967854466214, "tokens_seen": 2998403072 }, { "epoch": 0.03, "learning_rate": 0.0009837467921300258, "loss": 2.5705, "theoretical_loss": 3.3208856229512356, "tokens_seen": 2998534144 }, { "epoch": 0.03, "objective/train/docs_used": 1644380, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.542499542236328, "objective/train/theoretical_loss": 3.320880041937749, "objective/train/tokens_used": 28650976, "theoretical_loss": 3.320880041937749, "tokens_seen": 2998599680 }, { "epoch": 0.03, "learning_rate": 0.0009833190761334475, "loss": 2.5074, "theoretical_loss": 3.3208744610803898, "tokens_seen": 2998665216 }, { "epoch": 0.03, "learning_rate": 0.0009828913601368692, "loss": 2.6151, "theoretical_loss": 3.320863299834021, "tokens_seen": 2998796288 }, { "epoch": 0.03, "learning_rate": 0.0009824636441402908, "loss": 2.6994, "theoretical_loss": 3.320852139212068, "tokens_seen": 2998927360 }, { "epoch": 0.03, "learning_rate": 0.0009820359281437125, "loss": 2.568, "theoretical_loss": 3.3208409792144677, "tokens_seen": 2999058432 }, { "epoch": 0.03, "learning_rate": 0.0009816082121471344, "loss": 2.5552, "theoretical_loss": 3.320829819841158, "tokens_seen": 2999189504 }, { "epoch": 0.03, "learning_rate": 0.000981180496150556, "loss": 2.6719, "theoretical_loss": 3.320818661092077, "tokens_seen": 2999320576 }, { "epoch": 0.03, "learning_rate": 0.0009807527801539778, "loss": 2.5567, "theoretical_loss": 3.3208075029671624, "tokens_seen": 2999451648 }, { "epoch": 0.03, "learning_rate": 0.0009803250641573994, "loss": 2.5511, "theoretical_loss": 3.320796345466352, "tokens_seen": 2999582720 }, { "epoch": 0.03, "learning_rate": 0.0009798973481608211, "loss": 2.6608, "theoretical_loss": 3.320785188589584, "tokens_seen": 2999713792 }, { "epoch": 0.03, "learning_rate": 0.000979469632164243, "loss": 2.5947, "theoretical_loss": 3.3207740323367956, "tokens_seen": 2999844864 }, { "epoch": 0.03, "learning_rate": 0.0009790419161676647, "loss": 2.6511, "theoretical_loss": 3.3207628767079242, "tokens_seen": 2999975936 }, { "epoch": 0.03, "learning_rate": 0.0009786142001710863, "loss": 2.5573, "theoretical_loss": 3.3207517217029094, "tokens_seen": 3000107008 }, { "epoch": 0.03, "objective/train/docs_used": 1645056, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4763801097869873, "objective/train/theoretical_loss": 3.3207405673216877, "objective/train/tokens_used": 30289376, "theoretical_loss": 3.3207405673216877, "tokens_seen": 3000238080 }, { "epoch": 0.03, "learning_rate": 0.0009781864841745082, "loss": 2.6745, "theoretical_loss": 3.3207405673216877, "tokens_seen": 3000238080 }, { "epoch": 0.03, "learning_rate": 0.00097775876817793, "loss": 2.697, "theoretical_loss": 3.320729413564197, "tokens_seen": 3000369152 }, { "epoch": 0.03, "learning_rate": 0.0009773310521813516, "loss": 2.6853, "theoretical_loss": 3.3207182604303753, "tokens_seen": 3000500224 }, { "epoch": 0.03, "learning_rate": 0.0009769033361847733, "loss": 2.3445, "theoretical_loss": 3.320707107920161, "tokens_seen": 3000631296 }, { "epoch": 0.03, "learning_rate": 0.000976475620188195, "loss": 2.6763, "theoretical_loss": 3.3206959560334917, "tokens_seen": 3000762368 }, { "epoch": 0.03, "learning_rate": 0.0009760479041916168, "loss": 2.5198, "theoretical_loss": 3.320684804770305, "tokens_seen": 3000893440 }, { "epoch": 0.03, "learning_rate": 0.0009756201881950385, "loss": 2.7, "theoretical_loss": 3.3206736541305393, "tokens_seen": 3001024512 }, { "epoch": 0.03, "learning_rate": 0.0009751924721984602, "loss": 2.6958, "theoretical_loss": 3.3206625041141318, "tokens_seen": 3001155584 }, { "epoch": 0.04, "learning_rate": 0.000974764756201882, "loss": 2.6457, "theoretical_loss": 3.3206513547210212, "tokens_seen": 3001286656 }, { "epoch": 0.04, "learning_rate": 0.0009743370402053036, "loss": 2.7946, "theoretical_loss": 3.320640205951145, "tokens_seen": 3001417728 }, { "epoch": 0.04, "learning_rate": 0.0009739093242087254, "loss": 2.6682, "theoretical_loss": 3.3206290578044415, "tokens_seen": 3001548800 }, { "epoch": 0.04, "learning_rate": 0.0009734816082121472, "loss": 2.5484, "theoretical_loss": 3.3206179102808484, "tokens_seen": 3001679872 }, { "epoch": 0.04, "learning_rate": 0.0009730538922155689, "loss": 2.6724, "theoretical_loss": 3.3206067633803036, "tokens_seen": 3001810944 }, { "epoch": 0.04, "objective/train/docs_used": 1646327, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.547891855239868, "objective/train/theoretical_loss": 3.320601190163655, "objective/train/tokens_used": 31927776, "theoretical_loss": 3.320601190163655, "tokens_seen": 3001876480 }, { "epoch": 0.04, "learning_rate": 0.0009726261762189907, "loss": 2.5267, "theoretical_loss": 3.320595617102745, "tokens_seen": 3001942016 }, { "epoch": 0.04, "learning_rate": 0.0009721984602224123, "loss": 2.6548, "theoretical_loss": 3.320584471448111, "tokens_seen": 3002073088 }, { "epoch": 0.04, "learning_rate": 0.0009717707442258341, "loss": 2.5147, "theoretical_loss": 3.3205733264163393, "tokens_seen": 3002204160 }, { "epoch": 0.04, "learning_rate": 0.0009713430282292558, "loss": 2.4505, "theoretical_loss": 3.320562182007368, "tokens_seen": 3002335232 }, { "epoch": 0.04, "learning_rate": 0.0009709153122326775, "loss": 2.6305, "theoretical_loss": 3.320551038221135, "tokens_seen": 3002466304 }, { "epoch": 0.04, "learning_rate": 0.0009704875962360993, "loss": 2.5482, "theoretical_loss": 3.3205398950575784, "tokens_seen": 3002597376 }, { "epoch": 0.04, "learning_rate": 0.0009700598802395209, "loss": 2.7266, "theoretical_loss": 3.320528752516636, "tokens_seen": 3002728448 }, { "epoch": 0.04, "learning_rate": 0.0009696321642429427, "loss": 2.5155, "theoretical_loss": 3.3205176105982463, "tokens_seen": 3002859520 }, { "epoch": 0.04, "learning_rate": 0.0009692044482463645, "loss": 2.7628, "theoretical_loss": 3.320506469302347, "tokens_seen": 3002990592 }, { "epoch": 0.04, "learning_rate": 0.0009687767322497862, "loss": 2.6802, "theoretical_loss": 3.3204953286288763, "tokens_seen": 3003121664 }, { "epoch": 0.04, "learning_rate": 0.000968349016253208, "loss": 2.7921, "theoretical_loss": 3.3204841885777725, "tokens_seen": 3003252736 }, { "epoch": 0.04, "learning_rate": 0.0009679213002566296, "loss": 2.6088, "theoretical_loss": 3.3204730491489727, "tokens_seen": 3003383808 }, { "epoch": 0.04, "objective/train/docs_used": 1647543, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5402848720550537, "objective/train/theoretical_loss": 3.3204619103424164, "objective/train/tokens_used": 33566176, "theoretical_loss": 3.3204619103424164, "tokens_seen": 3003514880 }, { "epoch": 0.04, "learning_rate": 0.0009674935842600513, "loss": 2.6446, "theoretical_loss": 3.3204619103424164, "tokens_seen": 3003514880 }, { "epoch": 0.04, "learning_rate": 0.0009670658682634731, "loss": 2.5202, "theoretical_loss": 3.3204507721580403, "tokens_seen": 3003645952 }, { "epoch": 0.04, "learning_rate": 0.0009666381522668948, "loss": 2.4832, "theoretical_loss": 3.3204396345957834, "tokens_seen": 3003777024 }, { "epoch": 0.04, "learning_rate": 0.0009662104362703165, "loss": 2.6544, "theoretical_loss": 3.320428497655584, "tokens_seen": 3003908096 }, { "epoch": 0.04, "learning_rate": 0.0009657827202737382, "loss": 2.7507, "theoretical_loss": 3.320417361337379, "tokens_seen": 3004039168 }, { "epoch": 0.04, "learning_rate": 0.00096535500427716, "loss": 2.5786, "theoretical_loss": 3.3204062256411078, "tokens_seen": 3004170240 }, { "epoch": 0.04, "learning_rate": 0.0009649272882805818, "loss": 2.6649, "theoretical_loss": 3.320395090566708, "tokens_seen": 3004301312 }, { "epoch": 0.05, "learning_rate": 0.0009644995722840035, "loss": 2.6338, "theoretical_loss": 3.3203839561141173, "tokens_seen": 3004432384 }, { "epoch": 0.05, "learning_rate": 0.0009640718562874252, "loss": 2.664, "theoretical_loss": 3.320372822283275, "tokens_seen": 3004563456 }, { "epoch": 0.05, "learning_rate": 0.0009636441402908469, "loss": 2.6909, "theoretical_loss": 3.3203616890741183, "tokens_seen": 3004694528 }, { "epoch": 0.05, "learning_rate": 0.0009632164242942686, "loss": 2.4478, "theoretical_loss": 3.3203505564865856, "tokens_seen": 3004825600 }, { "epoch": 0.05, "learning_rate": 0.0009627887082976904, "loss": 2.6265, "theoretical_loss": 3.3203394245206153, "tokens_seen": 3004956672 }, { "epoch": 0.05, "learning_rate": 0.000962360992301112, "loss": 2.5755, "theoretical_loss": 3.320328293176145, "tokens_seen": 3005087744 }, { "epoch": 0.05, "objective/train/docs_used": 1648109, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5592408180236816, "objective/train/theoretical_loss": 3.3203227277369534, "objective/train/tokens_used": 35204576, "theoretical_loss": 3.3203227277369534, "tokens_seen": 3005153280 }, { "epoch": 0.05, "learning_rate": 0.0009619332763045337, "loss": 2.567, "theoretical_loss": 3.320317162453114, "tokens_seen": 3005218816 }, { "epoch": 0.05, "learning_rate": 0.0009615055603079555, "loss": 2.6035, "theoretical_loss": 3.3203060323514593, "tokens_seen": 3005349888 }, { "epoch": 0.05, "learning_rate": 0.0009610778443113773, "loss": 2.6209, "theoretical_loss": 3.3202949028711197, "tokens_seen": 3005480960 }, { "epoch": 0.05, "learning_rate": 0.0009606501283147991, "loss": 2.522, "theoretical_loss": 3.3202837740120335, "tokens_seen": 3005612032 }, { "epoch": 0.05, "learning_rate": 0.0009602224123182207, "loss": 2.5764, "theoretical_loss": 3.3202726457741387, "tokens_seen": 3005743104 }, { "epoch": 0.05, "learning_rate": 0.0009597946963216424, "loss": 2.6805, "theoretical_loss": 3.320261518157374, "tokens_seen": 3005874176 }, { "epoch": 0.05, "learning_rate": 0.0009593669803250642, "loss": 2.4835, "theoretical_loss": 3.3202503911616765, "tokens_seen": 3006005248 }, { "epoch": 0.05, "learning_rate": 0.0009589392643284859, "loss": 2.5907, "theoretical_loss": 3.320239264786986, "tokens_seen": 3006136320 }, { "epoch": 0.05, "learning_rate": 0.0009585115483319077, "loss": 2.617, "theoretical_loss": 3.3202281390332393, "tokens_seen": 3006267392 }, { "epoch": 0.05, "learning_rate": 0.0009580838323353293, "loss": 2.5027, "theoretical_loss": 3.320217013900376, "tokens_seen": 3006398464 }, { "epoch": 0.05, "learning_rate": 0.000957656116338751, "loss": 2.6857, "theoretical_loss": 3.3202058893883333, "tokens_seen": 3006529536 }, { "epoch": 0.05, "learning_rate": 0.0009572284003421729, "loss": 2.6411, "theoretical_loss": 3.3201947654970505, "tokens_seen": 3006660608 }, { "epoch": 0.05, "objective/train/docs_used": 1649212, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.7101857662200928, "objective/train/theoretical_loss": 3.320183642226465, "objective/train/tokens_used": 36842976, "theoretical_loss": 3.320183642226465, "tokens_seen": 3006791680 }, { "epoch": 0.05, "learning_rate": 0.0009568006843455946, "loss": 2.5369, "theoretical_loss": 3.320183642226465, "tokens_seen": 3006791680 }, { "epoch": 0.05, "learning_rate": 0.0009563729683490164, "loss": 2.5836, "theoretical_loss": 3.3201725195765155, "tokens_seen": 3006922752 }, { "epoch": 0.05, "learning_rate": 0.000955945252352438, "loss": 2.5358, "theoretical_loss": 3.3201613975471402, "tokens_seen": 3007053824 }, { "epoch": 0.05, "learning_rate": 0.0009555175363558597, "loss": 2.6035, "theoretical_loss": 3.3201502761382775, "tokens_seen": 3007184896 }, { "epoch": 0.05, "learning_rate": 0.0009550898203592815, "loss": 2.4594, "theoretical_loss": 3.320139155349866, "tokens_seen": 3007315968 }, { "epoch": 0.06, "learning_rate": 0.0009546621043627032, "loss": 2.6376, "theoretical_loss": 3.3201280351818436, "tokens_seen": 3007447040 }, { "epoch": 0.06, "learning_rate": 0.0009542343883661248, "loss": 2.5904, "theoretical_loss": 3.320116915634149, "tokens_seen": 3007578112 }, { "epoch": 0.06, "learning_rate": 0.0009538066723695466, "loss": 2.7616, "theoretical_loss": 3.3201057967067205, "tokens_seen": 3007709184 }, { "epoch": 0.06, "learning_rate": 0.0009533789563729683, "loss": 2.6076, "theoretical_loss": 3.3200946783994962, "tokens_seen": 3007840256 }, { "epoch": 0.06, "learning_rate": 0.0009529512403763902, "loss": 2.6121, "theoretical_loss": 3.3200835607124146, "tokens_seen": 3007971328 }, { "epoch": 0.06, "learning_rate": 0.0009525235243798119, "loss": 2.6083, "theoretical_loss": 3.3200724436454143, "tokens_seen": 3008102400 }, { "epoch": 0.06, "learning_rate": 0.0009520958083832335, "loss": 2.6831, "theoretical_loss": 3.3200613271984336, "tokens_seen": 3008233472 }, { "epoch": 0.06, "learning_rate": 0.0009516680923866553, "loss": 2.6156, "theoretical_loss": 3.3200502113714108, "tokens_seen": 3008364544 }, { "epoch": 0.06, "objective/train/docs_used": 1649940, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.545426368713379, "objective/train/theoretical_loss": 3.3200446536903643, "objective/train/tokens_used": 38481376, "theoretical_loss": 3.3200446536903643, "tokens_seen": 3008430080 }, { "epoch": 0.06, "learning_rate": 0.000951240376390077, "loss": 2.6849, "theoretical_loss": 3.3200390961642845, "tokens_seen": 3008495616 }, { "epoch": 0.06, "learning_rate": 0.0009508126603934988, "loss": 2.654, "theoretical_loss": 3.3200279815769926, "tokens_seen": 3008626688 }, { "epoch": 0.06, "learning_rate": 0.0009503849443969204, "loss": 2.6215, "theoretical_loss": 3.3200168676094743, "tokens_seen": 3008757760 }, { "epoch": 0.06, "learning_rate": 0.0009499572284003421, "loss": 2.4387, "theoretical_loss": 3.320005754261668, "tokens_seen": 3008888832 }, { "epoch": 0.06, "learning_rate": 0.0009495295124037639, "loss": 2.5621, "theoretical_loss": 3.319994641533511, "tokens_seen": 3009019904 }, { "epoch": 0.06, "learning_rate": 0.0009491017964071857, "loss": 2.731, "theoretical_loss": 3.319983529424943, "tokens_seen": 3009150976 }, { "epoch": 0.06, "learning_rate": 0.0009486740804106075, "loss": 2.5847, "theoretical_loss": 3.3199724179359027, "tokens_seen": 3009282048 }, { "epoch": 0.06, "learning_rate": 0.0009482463644140291, "loss": 2.5836, "theoretical_loss": 3.319961307066327, "tokens_seen": 3009413120 }, { "epoch": 0.06, "learning_rate": 0.0009478186484174508, "loss": 2.6488, "theoretical_loss": 3.3199501968161558, "tokens_seen": 3009544192 }, { "epoch": 0.06, "learning_rate": 0.0009473909324208726, "loss": 2.6244, "theoretical_loss": 3.319939087185327, "tokens_seen": 3009675264 }, { "epoch": 0.06, "learning_rate": 0.0009469632164242943, "loss": 2.6873, "theoretical_loss": 3.3199279781737796, "tokens_seen": 3009806336 }, { "epoch": 0.06, "learning_rate": 0.000946535500427716, "loss": 2.5921, "theoretical_loss": 3.3199168697814514, "tokens_seen": 3009937408 }, { "epoch": 0.06, "objective/train/docs_used": 1651249, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.582075834274292, "objective/train/theoretical_loss": 3.3199057620082812, "objective/train/tokens_used": 40119776, "theoretical_loss": 3.3199057620082812, "tokens_seen": 3010068480 }, { "epoch": 0.06, "learning_rate": 0.0009461077844311377, "loss": 2.5527, "theoretical_loss": 3.3199057620082812, "tokens_seen": 3010068480 }, { "epoch": 0.06, "learning_rate": 0.0009456800684345594, "loss": 2.7818, "theoretical_loss": 3.319894654854208, "tokens_seen": 3010199552 }, { "epoch": 0.06, "learning_rate": 0.0009452523524379812, "loss": 2.628, "theoretical_loss": 3.3198835483191695, "tokens_seen": 3010330624 }, { "epoch": 0.06, "learning_rate": 0.000944824636441403, "loss": 2.4649, "theoretical_loss": 3.319872442403105, "tokens_seen": 3010461696 }, { "epoch": 0.07, "learning_rate": 0.0009443969204448247, "loss": 2.7571, "theoretical_loss": 3.3198613371059524, "tokens_seen": 3010592768 }, { "epoch": 0.07, "learning_rate": 0.0009439692044482464, "loss": 2.5878, "theoretical_loss": 3.319850232427651, "tokens_seen": 3010723840 }, { "epoch": 0.07, "learning_rate": 0.0009435414884516681, "loss": 2.6015, "theoretical_loss": 3.3198391283681383, "tokens_seen": 3010854912 }, { "epoch": 0.07, "learning_rate": 0.0009431137724550899, "loss": 2.5823, "theoretical_loss": 3.3198280249273546, "tokens_seen": 3010985984 }, { "epoch": 0.07, "learning_rate": 0.0009426860564585116, "loss": 2.7222, "theoretical_loss": 3.319816922105237, "tokens_seen": 3011117056 }, { "epoch": 0.07, "learning_rate": 0.0009422583404619332, "loss": 2.6364, "theoretical_loss": 3.319805819901724, "tokens_seen": 3011248128 }, { "epoch": 0.07, "learning_rate": 0.000941830624465355, "loss": 2.6811, "theoretical_loss": 3.3197947183167553, "tokens_seen": 3011379200 }, { "epoch": 0.07, "learning_rate": 0.0009414029084687767, "loss": 2.8016, "theoretical_loss": 3.319783617350269, "tokens_seen": 3011510272 }, { "epoch": 0.07, "learning_rate": 0.0009409751924721985, "loss": 2.5657, "theoretical_loss": 3.319772517002204, "tokens_seen": 3011641344 }, { "epoch": 0.07, "objective/train/docs_used": 1651905, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 3.0052103996276855, "objective/train/theoretical_loss": 3.31976696706006, "objective/train/tokens_used": 41758176, "theoretical_loss": 3.31976696706006, "tokens_seen": 3011706880 }, { "epoch": 0.07, "learning_rate": 0.0009405474764756203, "loss": 2.6357, "theoretical_loss": 3.319761417272498, "tokens_seen": 3011772416 }, { "epoch": 0.07, "learning_rate": 0.0009401197604790419, "loss": 2.5734, "theoretical_loss": 3.319750318161091, "tokens_seen": 3011903488 }, { "epoch": 0.07, "learning_rate": 0.0009396920444824637, "loss": 2.424, "theoretical_loss": 3.3197392196679205, "tokens_seen": 3012034560 }, { "epoch": 0.07, "learning_rate": 0.0009392643284858854, "loss": 2.5344, "theoretical_loss": 3.3197281217929255, "tokens_seen": 3012165632 }, { "epoch": 0.07, "learning_rate": 0.0009388366124893071, "loss": 2.5689, "theoretical_loss": 3.319717024536045, "tokens_seen": 3012296704 }, { "epoch": 0.07, "learning_rate": 0.0009384088964927289, "loss": 2.4989, "theoretical_loss": 3.3197059278972176, "tokens_seen": 3012427776 }, { "epoch": 0.07, "learning_rate": 0.0009379811804961505, "loss": 2.6272, "theoretical_loss": 3.3196948318763817, "tokens_seen": 3012558848 }, { "epoch": 0.07, "learning_rate": 0.0009375534644995723, "loss": 2.5959, "theoretical_loss": 3.319683736473476, "tokens_seen": 3012689920 }, { "epoch": 0.07, "learning_rate": 0.000937125748502994, "loss": 2.5228, "theoretical_loss": 3.3196726416884395, "tokens_seen": 3012820992 }, { "epoch": 0.07, "learning_rate": 0.0009366980325064158, "loss": 2.7357, "theoretical_loss": 3.3196615475212106, "tokens_seen": 3012952064 }, { "epoch": 0.07, "learning_rate": 0.0009362703165098376, "loss": 2.6946, "theoretical_loss": 3.3196504539717284, "tokens_seen": 3013083136 }, { "epoch": 0.07, "learning_rate": 0.0009358426005132592, "loss": 2.7344, "theoretical_loss": 3.3196393610399317, "tokens_seen": 3013214208 }, { "epoch": 0.07, "objective/train/docs_used": 1652881, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.5316507816314697, "objective/train/theoretical_loss": 3.3196282687257583, "objective/train/tokens_used": 43396576, "theoretical_loss": 3.3196282687257583, "tokens_seen": 3013345280 }, { "epoch": 0.07, "learning_rate": 0.000935414884516681, "loss": 2.7259, "theoretical_loss": 3.3196282687257583, "tokens_seen": 3013345280 }, { "epoch": 0.07, "learning_rate": 0.0009349871685201027, "loss": 2.4882, "theoretical_loss": 3.3196171770291483, "tokens_seen": 3013476352 }, { "epoch": 0.07, "learning_rate": 0.0009345594525235244, "loss": 2.6169, "theoretical_loss": 3.3196060859500394, "tokens_seen": 3013607424 }, { "epoch": 0.08, "learning_rate": 0.0009341317365269461, "loss": 2.5293, "theoretical_loss": 3.319594995488371, "tokens_seen": 3013738496 }, { "epoch": 0.08, "learning_rate": 0.0009337040205303678, "loss": 2.7782, "theoretical_loss": 3.3195839056440812, "tokens_seen": 3013869568 }, { "epoch": 0.08, "learning_rate": 0.0009332763045337895, "loss": 2.6719, "theoretical_loss": 3.3195728164171094, "tokens_seen": 3014000640 }, { "epoch": 0.08, "learning_rate": 0.0009328485885372114, "loss": 2.7268, "theoretical_loss": 3.319561727807394, "tokens_seen": 3014131712 }, { "epoch": 0.08, "learning_rate": 0.0009324208725406331, "loss": 2.5997, "theoretical_loss": 3.3195506398148744, "tokens_seen": 3014262784 }, { "epoch": 0.08, "learning_rate": 0.0009319931565440548, "loss": 2.7602, "theoretical_loss": 3.319539552439489, "tokens_seen": 3014393856 }, { "epoch": 0.08, "learning_rate": 0.0009315654405474765, "loss": 2.5845, "theoretical_loss": 3.3195284656811763, "tokens_seen": 3014524928 }, { "epoch": 0.08, "learning_rate": 0.0009311377245508982, "loss": 2.7053, "theoretical_loss": 3.319517379539876, "tokens_seen": 3014656000 }, { "epoch": 0.08, "learning_rate": 0.00093071000855432, "loss": 2.6045, "theoretical_loss": 3.3195062940155258, "tokens_seen": 3014787072 }, { "epoch": 0.08, "learning_rate": 0.0009302822925577416, "loss": 2.6324, "theoretical_loss": 3.3194952091080654, "tokens_seen": 3014918144 }, { "epoch": 0.08, "objective/train/docs_used": 1653310, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.4821465015411377, "objective/train/theoretical_loss": 3.3194896668856497, "objective/train/tokens_used": 45034976, "theoretical_loss": 3.3194896668856497, "tokens_seen": 3014983680 }, { "epoch": 0.08, "learning_rate": 0.0009298545765611634, "loss": 2.532, "theoretical_loss": 3.3194841248174334, "tokens_seen": 3015049216 }, { "epoch": 0.08, "learning_rate": 0.0009294268605645851, "loss": 2.544, "theoretical_loss": 3.3194730411435684, "tokens_seen": 3015180288 }, { "epoch": 0.08, "learning_rate": 0.0009289991445680068, "loss": 2.826, "theoretical_loss": 3.3194619580864098, "tokens_seen": 3015311360 }, { "epoch": 0.08, "learning_rate": 0.0009285714285714287, "loss": 2.7561, "theoretical_loss": 3.3194508756458965, "tokens_seen": 3015442432 }, { "epoch": 0.08, "learning_rate": 0.0009281437125748503, "loss": 2.5692, "theoretical_loss": 3.319439793821967, "tokens_seen": 3015573504 }, { "epoch": 0.08, "learning_rate": 0.000927715996578272, "loss": 2.6322, "theoretical_loss": 3.3194287126145596, "tokens_seen": 3015704576 }, { "epoch": 0.08, "learning_rate": 0.0009272882805816938, "loss": 2.6346, "theoretical_loss": 3.3194176320236144, "tokens_seen": 3015835648 }, { "epoch": 0.08, "learning_rate": 0.0009268605645851155, "loss": 2.7908, "theoretical_loss": 3.31940655204907, "tokens_seen": 3015966720 }, { "epoch": 0.08, "learning_rate": 0.0009264328485885373, "loss": 2.6439, "theoretical_loss": 3.319395472690865, "tokens_seen": 3016097792 }, { "epoch": 0.08, "learning_rate": 0.0009260051325919589, "loss": 2.5589, "theoretical_loss": 3.3193843939489382, "tokens_seen": 3016228864 }, { "epoch": 0.08, "learning_rate": 0.0009255774165953806, "loss": 2.6926, "theoretical_loss": 3.319373315823229, "tokens_seen": 3016359936 }, { "epoch": 0.08, "learning_rate": 0.0009251497005988024, "loss": 2.7646, "theoretical_loss": 3.3193622383136763, "tokens_seen": 3016491008 }, { "epoch": 0.08, "objective/train/docs_used": 1654644, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.3390886783599854, "objective/train/theoretical_loss": 3.3193511614202187, "objective/train/tokens_used": 46673376, "theoretical_loss": 3.3193511614202187, "tokens_seen": 3016622080 }, { "epoch": 0.08, "learning_rate": 0.0009247219846022242, "loss": 2.5287, "theoretical_loss": 3.3193511614202187, "tokens_seen": 3016622080 }, { "epoch": 0.09, "learning_rate": 0.000924294268605646, "loss": 2.7607, "theoretical_loss": 3.319340085142796, "tokens_seen": 3016753152 }, { "epoch": 0.09, "learning_rate": 0.0009238665526090676, "loss": 2.6289, "theoretical_loss": 3.319329009481346, "tokens_seen": 3016884224 }, { "epoch": 0.09, "learning_rate": 0.0009234388366124893, "loss": 2.7648, "theoretical_loss": 3.3193179344358086, "tokens_seen": 3017015296 }, { "epoch": 0.09, "learning_rate": 0.0009230111206159111, "loss": 2.497, "theoretical_loss": 3.319306860006122, "tokens_seen": 3017146368 }, { "epoch": 0.09, "learning_rate": 0.0009225834046193328, "loss": 2.4963, "theoretical_loss": 3.319295786192226, "tokens_seen": 3017277440 }, { "epoch": 0.09, "learning_rate": 0.0009221556886227545, "loss": 2.5823, "theoretical_loss": 3.319284712994059, "tokens_seen": 3017408512 }, { "epoch": 0.09, "learning_rate": 0.0009217279726261762, "loss": 2.6555, "theoretical_loss": 3.3192736404115606, "tokens_seen": 3017539584 }, { "epoch": 0.09, "learning_rate": 0.0009213002566295979, "loss": 2.5264, "theoretical_loss": 3.3192625684446693, "tokens_seen": 3017670656 }, { "epoch": 0.09, "learning_rate": 0.0009208725406330197, "loss": 2.397, "theoretical_loss": 3.3192514970933242, "tokens_seen": 3017801728 }, { "epoch": 0.09, "learning_rate": 0.0009204448246364415, "loss": 2.6273, "theoretical_loss": 3.319240426357465, "tokens_seen": 3017932800 }, { "epoch": 0.09, "learning_rate": 0.0009200171086398631, "loss": 2.6414, "theoretical_loss": 3.31922935623703, "tokens_seen": 3018063872 }, { "epoch": 0.09, "learning_rate": 0.0009195893926432849, "loss": 2.5706, "theoretical_loss": 3.3192182867319584, "tokens_seen": 3018194944 }, { "epoch": 0.09, "objective/train/docs_used": 1655335, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.3051180839538574, "objective/train/theoretical_loss": 3.319212752210165, "objective/train/tokens_used": 48311776, "theoretical_loss": 3.319212752210165, "tokens_seen": 3018260480 }, { "epoch": 0.09, "learning_rate": 0.0009191616766467066, "loss": 2.5393, "theoretical_loss": 3.3192072178421896, "tokens_seen": 3018326016 }, { "epoch": 0.09, "learning_rate": 0.0009187339606501284, "loss": 2.6545, "theoretical_loss": 3.319196149567662, "tokens_seen": 3018457088 }, { "epoch": 0.09, "learning_rate": 0.00091830624465355, "loss": 2.5623, "theoretical_loss": 3.3191850819083157, "tokens_seen": 3018588160 }, { "epoch": 0.09, "learning_rate": 0.0009178785286569717, "loss": 2.6804, "theoretical_loss": 3.319174014864089, "tokens_seen": 3018719232 }, { "epoch": 0.09, "learning_rate": 0.0009174508126603935, "loss": 2.8051, "theoretical_loss": 3.319162948434921, "tokens_seen": 3018850304 }, { "epoch": 0.09, "learning_rate": 0.0009170230966638152, "loss": 2.7044, "theoretical_loss": 3.319151882620752, "tokens_seen": 3018981376 }, { "epoch": 0.09, "learning_rate": 0.0009165953806672371, "loss": 2.5421, "theoretical_loss": 3.3191408174215193, "tokens_seen": 3019112448 }, { "epoch": 0.09, "learning_rate": 0.0009161676646706587, "loss": 2.7474, "theoretical_loss": 3.3191297528371635, "tokens_seen": 3019243520 }, { "epoch": 0.09, "learning_rate": 0.0009157399486740804, "loss": 2.5804, "theoretical_loss": 3.319118688867623, "tokens_seen": 3019374592 }, { "epoch": 0.09, "learning_rate": 0.0009153122326775022, "loss": 2.5145, "theoretical_loss": 3.319107625512837, "tokens_seen": 3019505664 }, { "epoch": 0.09, "learning_rate": 0.0009148845166809239, "loss": 2.746, "theoretical_loss": 3.3190965627727445, "tokens_seen": 3019636736 }, { "epoch": 0.09, "learning_rate": 0.0009144568006843457, "loss": 2.5949, "theoretical_loss": 3.3190855006472857, "tokens_seen": 3019767808 }, { "epoch": 0.09, "objective/train/docs_used": 1656670, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.1859257221221924, "objective/train/theoretical_loss": 3.3190744391363984, "objective/train/tokens_used": 49950176, "theoretical_loss": 3.3190744391363984, "tokens_seen": 3019898880 }, { "epoch": 0.1, "learning_rate": 0.0009140290846877673, "loss": 2.6334, "theoretical_loss": 3.3190744391363984, "tokens_seen": 3019898880 }, { "epoch": 0.1, "learning_rate": 0.000913601368691189, "loss": 2.6976, "theoretical_loss": 3.3190633782400223, "tokens_seen": 3020029952 }, { "epoch": 0.1, "learning_rate": 0.0009131736526946108, "loss": 2.5658, "theoretical_loss": 3.3190523179580973, "tokens_seen": 3020161024 }, { "epoch": 0.1, "learning_rate": 0.0009127459366980325, "loss": 2.7495, "theoretical_loss": 3.3190412582905617, "tokens_seen": 3020292096 }, { "epoch": 0.1, "learning_rate": 0.0009123182207014543, "loss": 2.7093, "theoretical_loss": 3.319030199237355, "tokens_seen": 3020423168 }, { "epoch": 0.1, "learning_rate": 0.000911890504704876, "loss": 2.5295, "theoretical_loss": 3.3190191407984164, "tokens_seen": 3020554240 }, { "epoch": 0.1, "learning_rate": 0.0009114627887082977, "loss": 2.6081, "theoretical_loss": 3.3190080829736854, "tokens_seen": 3020685312 }, { "epoch": 0.1, "learning_rate": 0.0009110350727117195, "loss": 2.6166, "theoretical_loss": 3.318997025763101, "tokens_seen": 3020816384 }, { "epoch": 0.1, "learning_rate": 0.0009106073567151412, "loss": 2.7691, "theoretical_loss": 3.318985969166602, "tokens_seen": 3020947456 }, { "epoch": 0.1, "learning_rate": 0.0009101796407185628, "loss": 2.6465, "theoretical_loss": 3.3189749131841286, "tokens_seen": 3021078528 }, { "epoch": 0.1, "learning_rate": 0.0009097519247219846, "loss": 2.5314, "theoretical_loss": 3.3189638578156195, "tokens_seen": 3021209600 }, { "epoch": 0.1, "learning_rate": 0.0009093242087254063, "loss": 2.8726, "theoretical_loss": 3.3189528030610136, "tokens_seen": 3021340672 }, { "epoch": 0.1, "learning_rate": 0.0009088964927288281, "loss": 2.6762, "theoretical_loss": 3.318941748920251, "tokens_seen": 3021471744 }, { "epoch": 0.1, "objective/train/docs_used": 1657192, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.9018635749816895, "objective/train/theoretical_loss": 3.318936222080042, "objective/train/tokens_used": 51588576, "theoretical_loss": 3.318936222080042, "tokens_seen": 3021537280 }, { "epoch": 0.1, "learning_rate": 0.0009084687767322499, "loss": 2.643, "theoretical_loss": 3.318930695393271, "tokens_seen": 3021602816 }, { "epoch": 0.1, "learning_rate": 0.0009080410607356715, "loss": 2.6665, "theoretical_loss": 3.3189196424800116, "tokens_seen": 3021733888 }, { "epoch": 0.1, "learning_rate": 0.0009076133447390933, "loss": 2.6265, "theoretical_loss": 3.3189085901804134, "tokens_seen": 3021864960 }, { "epoch": 0.1, "learning_rate": 0.000907185628742515, "loss": 2.5788, "theoretical_loss": 3.3188975384944155, "tokens_seen": 3021996032 }, { "epoch": 0.1, "learning_rate": 0.0009067579127459367, "loss": 2.6531, "theoretical_loss": 3.318886487421957, "tokens_seen": 3022127104 }, { "epoch": 0.1, "learning_rate": 0.0009063301967493585, "loss": 2.7177, "theoretical_loss": 3.318875436962977, "tokens_seen": 3022258176 }, { "epoch": 0.1, "learning_rate": 0.0009059024807527801, "loss": 2.4989, "theoretical_loss": 3.3188643871174155, "tokens_seen": 3022389248 }, { "epoch": 0.1, "learning_rate": 0.0009054747647562019, "loss": 2.5249, "theoretical_loss": 3.318853337885211, "tokens_seen": 3022520320 }, { "epoch": 0.1, "learning_rate": 0.0009050470487596236, "loss": 2.6031, "theoretical_loss": 3.318842289266304, "tokens_seen": 3022651392 }, { "epoch": 0.1, "learning_rate": 0.0009046193327630453, "loss": 2.5702, "theoretical_loss": 3.3188312412606327, "tokens_seen": 3022782464 }, { "epoch": 0.1, "learning_rate": 0.0009041916167664672, "loss": 2.5348, "theoretical_loss": 3.3188201938681368, "tokens_seen": 3022913536 }, { "epoch": 0.11, "learning_rate": 0.0009037639007698888, "loss": 2.5876, "theoretical_loss": 3.318809147088756, "tokens_seen": 3023044608 }, { "epoch": 0.11, "objective/train/docs_used": 1658380, "objective/train/instantaneous_batch_size": 16, "objective/train/instantaneous_microbatch_size": 16384, "objective/train/original_loss": 2.8608322143554688, "objective/train/theoretical_loss": 3.3187981009224297, "objective/train/tokens_used": 53226976, "theoretical_loss": 3.3187981009224297, "tokens_seen": 3023175680 }, { "epoch": 0.11, "learning_rate": 0.0009033361847733106, "loss": 2.6644, "theoretical_loss": 3.3187981009224297, "tokens_seen": 3023175680 }, { "epoch": 0.11, "learning_rate": 0.0009029084687767323, "loss": 2.5824, "theoretical_loss": 3.3187870553690972, "tokens_seen": 3023306752 } ], "max_steps": 2362, "num_train_epochs": 9223372036854775807, "total_flos": 1.6789580808192e+16, "trial_name": null, "trial_params": null }