| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.993258426966292, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0299625468164794, | |
| "grad_norm": 2.159330801984932, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7997, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0599250936329588, | |
| "grad_norm": 2.575219453584873, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7216, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0898876404494382, | |
| "grad_norm": 0.9472938732683952, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7072, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1198501872659176, | |
| "grad_norm": 0.9037652307729707, | |
| "learning_rate": 5e-06, | |
| "loss": 0.686, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.149812734082397, | |
| "grad_norm": 0.9332908648588308, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6654, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1797752808988764, | |
| "grad_norm": 2.546264321590982, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6592, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.20973782771535582, | |
| "grad_norm": 0.8963518726180082, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6539, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2397003745318352, | |
| "grad_norm": 0.6474151088056208, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6484, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2696629213483146, | |
| "grad_norm": 0.7184102062533572, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6455, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.299625468164794, | |
| "grad_norm": 0.7912271314589597, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6422, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3295880149812734, | |
| "grad_norm": 0.5702593280439339, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6335, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3595505617977528, | |
| "grad_norm": 0.6067788640506643, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6252, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3895131086142322, | |
| "grad_norm": 0.7543519184412004, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6278, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.41947565543071164, | |
| "grad_norm": 1.3158399586116993, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6344, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.449438202247191, | |
| "grad_norm": 0.9280984142139298, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6261, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4794007490636704, | |
| "grad_norm": 0.8050698415939164, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6177, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5093632958801498, | |
| "grad_norm": 0.6632585021759287, | |
| "learning_rate": 5e-06, | |
| "loss": 0.622, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5393258426966292, | |
| "grad_norm": 0.7452935828094429, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6223, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5692883895131086, | |
| "grad_norm": 0.5259568899134429, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6183, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.599250936329588, | |
| "grad_norm": 1.0908063668428212, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6164, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6292134831460674, | |
| "grad_norm": 0.5904142262846527, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6247, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6591760299625468, | |
| "grad_norm": 0.48963729877546575, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6097, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6891385767790262, | |
| "grad_norm": 0.5647584677724115, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6193, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7191011235955056, | |
| "grad_norm": 0.6352290021696486, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6183, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7490636704119851, | |
| "grad_norm": 0.5114633845114385, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6163, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7790262172284644, | |
| "grad_norm": 0.7063368471913241, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6093, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8089887640449438, | |
| "grad_norm": 0.9028219544074879, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6135, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8389513108614233, | |
| "grad_norm": 0.6542303935292434, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6146, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8689138576779026, | |
| "grad_norm": 0.9490011650791124, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6258, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.898876404494382, | |
| "grad_norm": 0.5141275957416789, | |
| "learning_rate": 5e-06, | |
| "loss": 0.608, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9288389513108615, | |
| "grad_norm": 0.4695093914592938, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6086, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9588014981273408, | |
| "grad_norm": 0.4886269533641591, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6055, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9887640449438202, | |
| "grad_norm": 0.4785326651206929, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6122, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9977528089887641, | |
| "eval_loss": 0.6095167398452759, | |
| "eval_runtime": 180.8435, | |
| "eval_samples_per_second": 49.723, | |
| "eval_steps_per_second": 0.393, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0187265917602997, | |
| "grad_norm": 0.8136150743659245, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5787, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.048689138576779, | |
| "grad_norm": 0.5667433740151928, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5582, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0786516853932584, | |
| "grad_norm": 0.5036359850721361, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5638, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1086142322097379, | |
| "grad_norm": 0.5623087149261949, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5599, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1385767790262173, | |
| "grad_norm": 0.488809840644991, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5619, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1685393258426966, | |
| "grad_norm": 0.7591237017789003, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5617, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.198501872659176, | |
| "grad_norm": 0.7394067985885456, | |
| "learning_rate": 5e-06, | |
| "loss": 0.554, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2284644194756553, | |
| "grad_norm": 0.6131933237418792, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5646, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2584269662921348, | |
| "grad_norm": 0.554150130587659, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5653, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2883895131086143, | |
| "grad_norm": 0.5751376390924479, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5581, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3183520599250937, | |
| "grad_norm": 0.5120666058669939, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5651, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.348314606741573, | |
| "grad_norm": 0.8368555832046994, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5669, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3782771535580525, | |
| "grad_norm": 0.7407992028160174, | |
| "learning_rate": 5e-06, | |
| "loss": 0.554, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4082397003745317, | |
| "grad_norm": 0.5156759371159569, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5588, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4382022471910112, | |
| "grad_norm": 0.49080996761818096, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5632, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4681647940074907, | |
| "grad_norm": 0.4910042182094872, | |
| "learning_rate": 5e-06, | |
| "loss": 0.561, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.4981273408239701, | |
| "grad_norm": 0.5255273950611783, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5605, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5280898876404494, | |
| "grad_norm": 0.46327981869558943, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5618, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5580524344569289, | |
| "grad_norm": 0.55012055750815, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5568, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5880149812734081, | |
| "grad_norm": 0.5105338531939311, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5672, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6179775280898876, | |
| "grad_norm": 0.5336419797454037, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5593, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.647940074906367, | |
| "grad_norm": 0.9436464610139725, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5608, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6779026217228465, | |
| "grad_norm": 0.6016864999378152, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5588, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7078651685393258, | |
| "grad_norm": 0.5010354324689145, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5548, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7378277153558053, | |
| "grad_norm": 0.4589724783243399, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5665, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7677902621722845, | |
| "grad_norm": 0.5015630178996147, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5636, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.797752808988764, | |
| "grad_norm": 0.5495613719565868, | |
| "learning_rate": 5e-06, | |
| "loss": 0.565, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8277153558052435, | |
| "grad_norm": 0.5539975776071888, | |
| "learning_rate": 5e-06, | |
| "loss": 0.562, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.857677902621723, | |
| "grad_norm": 0.5053725200868951, | |
| "learning_rate": 5e-06, | |
| "loss": 0.551, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8876404494382022, | |
| "grad_norm": 0.4543290751635621, | |
| "learning_rate": 5e-06, | |
| "loss": 0.556, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9176029962546817, | |
| "grad_norm": 0.47157192643041534, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5602, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.947565543071161, | |
| "grad_norm": 0.4784340330073252, | |
| "learning_rate": 5e-06, | |
| "loss": 0.557, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.9775280898876404, | |
| "grad_norm": 0.44605574199738396, | |
| "learning_rate": 5e-06, | |
| "loss": 0.562, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.9985018726591761, | |
| "eval_loss": 0.6012518405914307, | |
| "eval_runtime": 181.0833, | |
| "eval_samples_per_second": 49.657, | |
| "eval_steps_per_second": 0.392, | |
| "step": 667 | |
| }, | |
| { | |
| "epoch": 2.00749063670412, | |
| "grad_norm": 0.8623975331438202, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5432, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0374531835205993, | |
| "grad_norm": 0.6324789071436193, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5019, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.067415730337079, | |
| "grad_norm": 0.5758535175615167, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5099, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.097378277153558, | |
| "grad_norm": 0.6234430093296897, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5112, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1273408239700373, | |
| "grad_norm": 0.5881227652440947, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5022, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.157303370786517, | |
| "grad_norm": 0.6192814926150049, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5066, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.1872659176029963, | |
| "grad_norm": 0.5117435754957025, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5064, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2172284644194757, | |
| "grad_norm": 0.5619958642740768, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5046, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.247191011235955, | |
| "grad_norm": 0.8267859788370541, | |
| "learning_rate": 5e-06, | |
| "loss": 0.511, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.2771535580524347, | |
| "grad_norm": 0.7317931221994743, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5092, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3071161048689137, | |
| "grad_norm": 0.6283385492658163, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5133, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.337078651685393, | |
| "grad_norm": 0.5246018099952993, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5131, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.3670411985018727, | |
| "grad_norm": 0.5366582911908819, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5163, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.397003745318352, | |
| "grad_norm": 0.5455507417027214, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5174, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4269662921348316, | |
| "grad_norm": 0.47939087317987383, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5117, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4569288389513106, | |
| "grad_norm": 0.5848286494949987, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5101, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.48689138576779, | |
| "grad_norm": 0.5885231295942317, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5098, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5168539325842696, | |
| "grad_norm": 0.49238621301397356, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5128, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.546816479400749, | |
| "grad_norm": 0.5319632499579365, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5129, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5767790262172285, | |
| "grad_norm": 0.5235625882031714, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5111, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.606741573033708, | |
| "grad_norm": 0.5063693847116817, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5125, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.6367041198501875, | |
| "grad_norm": 0.5425085022408588, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5067, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.49758835042040306, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5112, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.696629213483146, | |
| "grad_norm": 0.5128503684055458, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5163, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7265917602996255, | |
| "grad_norm": 0.5751048660393648, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5104, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.756554307116105, | |
| "grad_norm": 0.6107618204066423, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5124, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.7865168539325844, | |
| "grad_norm": 0.4918065441260285, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5062, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8164794007490634, | |
| "grad_norm": 0.49772220806864265, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5119, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.846441947565543, | |
| "grad_norm": 0.560475095793381, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5134, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8764044943820224, | |
| "grad_norm": 0.5127175297281757, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5117, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.906367041198502, | |
| "grad_norm": 0.6130240398398701, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5188, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.9363295880149813, | |
| "grad_norm": 0.6513015524907453, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5102, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.966292134831461, | |
| "grad_norm": 0.5631449617834976, | |
| "learning_rate": 5e-06, | |
| "loss": 0.517, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.993258426966292, | |
| "eval_loss": 0.605403482913971, | |
| "eval_runtime": 181.5085, | |
| "eval_samples_per_second": 49.54, | |
| "eval_steps_per_second": 0.391, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 2.993258426966292, | |
| "step": 999, | |
| "total_flos": 1672943448883200.0, | |
| "train_loss": 0.5696583624716636, | |
| "train_runtime": 30032.5257, | |
| "train_samples_per_second": 17.065, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 999, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1672943448883200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |