| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9953632148377123, | |
| "eval_steps": 500, | |
| "global_step": 969, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.030911901081916538, | |
| "grad_norm": 2.0187587823402606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9224, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.061823802163833076, | |
| "grad_norm": 5.23374455774046, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8169, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09273570324574962, | |
| "grad_norm": 1.1567925380600657, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7799, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12364760432766615, | |
| "grad_norm": 1.3555247489499298, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7536, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1545595054095827, | |
| "grad_norm": 0.8540206480635643, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7412, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18547140649149924, | |
| "grad_norm": 1.0411208272577643, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7188, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21638330757341576, | |
| "grad_norm": 0.772135725000269, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7099, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2472952086553323, | |
| "grad_norm": 0.718962153128519, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7044, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2782071097372488, | |
| "grad_norm": 0.9018851336703633, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7042, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3091190108191654, | |
| "grad_norm": 0.5501975291936315, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6898, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3400309119010819, | |
| "grad_norm": 0.8644622620986302, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6901, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.37094281298299847, | |
| "grad_norm": 0.5948602458220357, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6828, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.401854714064915, | |
| "grad_norm": 0.720162590208508, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6858, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4327666151468315, | |
| "grad_norm": 0.5833952154939747, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6745, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46367851622874806, | |
| "grad_norm": 0.6768653844503261, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6764, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4945904173106646, | |
| "grad_norm": 0.5707017440465438, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6734, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5255023183925811, | |
| "grad_norm": 0.52632417149709, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6746, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5564142194744977, | |
| "grad_norm": 0.5660566537871714, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6654, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5873261205564142, | |
| "grad_norm": 0.6150804490960232, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6669, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6182380216383307, | |
| "grad_norm": 0.5523983219898584, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6722, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6491499227202473, | |
| "grad_norm": 0.7211222999715671, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6689, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6800618238021638, | |
| "grad_norm": 0.7196378524123959, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6685, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7109737248840804, | |
| "grad_norm": 0.7400759740386526, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6637, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7418856259659969, | |
| "grad_norm": 0.5515081055008348, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6602, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7727975270479135, | |
| "grad_norm": 0.5306297630685508, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6641, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.80370942812983, | |
| "grad_norm": 0.5220389228174283, | |
| "learning_rate": 5e-06, | |
| "loss": 0.657, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8346213292117465, | |
| "grad_norm": 0.5736057539218636, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6606, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.865533230293663, | |
| "grad_norm": 0.6831384912164588, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6604, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8964451313755796, | |
| "grad_norm": 0.6555441815061606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6617, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9273570324574961, | |
| "grad_norm": 0.5107130356234296, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6577, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9582689335394127, | |
| "grad_norm": 0.5294921581221994, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6554, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9891808346213292, | |
| "grad_norm": 0.5933706977854671, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6662, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9984544049459042, | |
| "eval_loss": 0.6590365171432495, | |
| "eval_runtime": 113.7467, | |
| "eval_samples_per_second": 76.618, | |
| "eval_steps_per_second": 0.607, | |
| "step": 323 | |
| }, | |
| { | |
| "epoch": 1.0200927357032457, | |
| "grad_norm": 0.6466092267278452, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6283, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0510046367851622, | |
| "grad_norm": 0.5455963224340917, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6142, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0819165378670788, | |
| "grad_norm": 0.5457563493460516, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6136, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1128284389489953, | |
| "grad_norm": 0.5123524043102256, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6186, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1437403400309119, | |
| "grad_norm": 0.6133162659271869, | |
| "learning_rate": 5e-06, | |
| "loss": 0.612, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1746522411128284, | |
| "grad_norm": 0.531392662924999, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6093, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.205564142194745, | |
| "grad_norm": 0.5434898390306484, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6031, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2364760432766615, | |
| "grad_norm": 0.665502589919703, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6153, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.267387944358578, | |
| "grad_norm": 0.5263248988074614, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6154, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2982998454404946, | |
| "grad_norm": 0.6348966883540149, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6112, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3292117465224111, | |
| "grad_norm": 0.5230511187445398, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6121, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3601236476043277, | |
| "grad_norm": 0.5500625021812695, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6109, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3910355486862442, | |
| "grad_norm": 0.7153093571149675, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6123, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4219474497681608, | |
| "grad_norm": 0.6382419366334218, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6188, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4528593508500773, | |
| "grad_norm": 0.5471481346135804, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6179, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4837712519319939, | |
| "grad_norm": 0.5409301526875265, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6121, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5146831530139102, | |
| "grad_norm": 0.5070509983871482, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6081, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.545595054095827, | |
| "grad_norm": 0.6049383227511866, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6111, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5765069551777433, | |
| "grad_norm": 0.5173228367317094, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6088, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.60741885625966, | |
| "grad_norm": 0.5087487433614813, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6077, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6383307573415764, | |
| "grad_norm": 0.6681086881393504, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6073, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6692426584234932, | |
| "grad_norm": 0.9262470601774611, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6088, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.7001545595054095, | |
| "grad_norm": 0.5528042557953681, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6163, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7310664605873263, | |
| "grad_norm": 0.5064527021504606, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6064, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7619783616692426, | |
| "grad_norm": 0.5307893550951357, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6132, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7928902627511591, | |
| "grad_norm": 0.5363288571581243, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6142, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8238021638330757, | |
| "grad_norm": 0.4985720023183297, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6176, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8547140649149922, | |
| "grad_norm": 0.5256325470444526, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6126, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8856259659969088, | |
| "grad_norm": 0.6274813396972104, | |
| "learning_rate": 5e-06, | |
| "loss": 0.612, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9165378670788253, | |
| "grad_norm": 0.5789315780747083, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6125, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.947449768160742, | |
| "grad_norm": 0.5907958566056928, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6062, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9783616692426584, | |
| "grad_norm": 0.5292904118235623, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6121, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6483988761901855, | |
| "eval_runtime": 114.5876, | |
| "eval_samples_per_second": 76.055, | |
| "eval_steps_per_second": 0.602, | |
| "step": 647 | |
| }, | |
| { | |
| "epoch": 2.009273570324575, | |
| "grad_norm": 1.0140873239994894, | |
| "learning_rate": 5e-06, | |
| "loss": 0.597, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0401854714064913, | |
| "grad_norm": 0.630334813080006, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5614, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.071097372488408, | |
| "grad_norm": 0.7038598022498512, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5613, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.1020092735703244, | |
| "grad_norm": 0.625422416199728, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5652, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.132921174652241, | |
| "grad_norm": 0.6504517407003202, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5646, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.1638330757341575, | |
| "grad_norm": 0.6005779084691315, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5635, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1947449768160743, | |
| "grad_norm": 0.5153825857145766, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5642, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.2256568778979906, | |
| "grad_norm": 0.6609026821979475, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5636, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.2565687789799074, | |
| "grad_norm": 0.7763886247938674, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5646, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2874806800618237, | |
| "grad_norm": 0.5940744564359401, | |
| "learning_rate": 5e-06, | |
| "loss": 0.566, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.3183925811437405, | |
| "grad_norm": 0.8987915044882564, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5639, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.349304482225657, | |
| "grad_norm": 0.7853307481543791, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5652, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3802163833075736, | |
| "grad_norm": 0.5555357862981527, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5695, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.41112828438949, | |
| "grad_norm": 0.5858397000016529, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5674, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.4420401854714067, | |
| "grad_norm": 0.5809649704126684, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5666, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.472952086553323, | |
| "grad_norm": 0.6079171026499909, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5662, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5038639876352393, | |
| "grad_norm": 0.6460104668614653, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5724, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.534775888717156, | |
| "grad_norm": 0.5036700308469926, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5689, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.565687789799073, | |
| "grad_norm": 0.5720133645014428, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5694, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.596599690880989, | |
| "grad_norm": 0.4978320121327588, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5688, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.6275115919629055, | |
| "grad_norm": 0.5742589206659079, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5677, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.6584234930448223, | |
| "grad_norm": 0.8122081568789994, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5685, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.689335394126739, | |
| "grad_norm": 0.5964179415080593, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5695, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.7202472952086554, | |
| "grad_norm": 0.5560894480978509, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5718, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.7511591962905717, | |
| "grad_norm": 0.5757117113268359, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5696, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7820710973724885, | |
| "grad_norm": 0.5435787690358811, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5676, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.812982998454405, | |
| "grad_norm": 0.6300976369973321, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5725, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.8438948995363216, | |
| "grad_norm": 0.6341460273574823, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5699, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.874806800618238, | |
| "grad_norm": 0.6961502333247155, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5694, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.9057187017001547, | |
| "grad_norm": 0.5370380886104864, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5712, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.936630602782071, | |
| "grad_norm": 0.595358378564684, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5691, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.9675425038639878, | |
| "grad_norm": 0.5413573519766721, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5679, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9953632148377123, | |
| "eval_loss": 0.6511133909225464, | |
| "eval_runtime": 110.2332, | |
| "eval_samples_per_second": 79.06, | |
| "eval_steps_per_second": 0.626, | |
| "step": 969 | |
| }, | |
| { | |
| "epoch": 2.9953632148377123, | |
| "step": 969, | |
| "total_flos": 1622692331520000.0, | |
| "train_loss": 0.6248709833166793, | |
| "train_runtime": 16382.8374, | |
| "train_samples_per_second": 30.32, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 969, | |
| "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": 1622692331520000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |