| { | |
| "best_metric": 37.93360828750764, | |
| "best_model_checkpoint": "/kaggle/working/whisper-small-fa/checkpoint-1200", | |
| "epoch": 0.5862237420615535, | |
| "eval_steps": 200, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004885197850512946, | |
| "grad_norm": Infinity, | |
| "learning_rate": 0.0, | |
| "loss": 1.0812, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004885197850512946, | |
| "grad_norm": 7.0439229011535645, | |
| "learning_rate": 1.6e-07, | |
| "loss": 1.1303, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009770395701025891, | |
| "grad_norm": 6.462230205535889, | |
| "learning_rate": 3.6e-07, | |
| "loss": 1.1685, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014655593551538837, | |
| "grad_norm": 6.182502269744873, | |
| "learning_rate": 5.6e-07, | |
| "loss": 1.1928, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.019540791402051783, | |
| "grad_norm": 5.509102821350098, | |
| "learning_rate": 7.6e-07, | |
| "loss": 1.0699, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.024425989252564728, | |
| "grad_norm": 6.310844421386719, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 1.1117, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.029311187103077674, | |
| "grad_norm": 4.859102725982666, | |
| "learning_rate": 1.1600000000000001e-06, | |
| "loss": 1.0681, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03419638495359062, | |
| "grad_norm": 5.078814506530762, | |
| "learning_rate": 1.3600000000000001e-06, | |
| "loss": 1.0524, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.039081582804103565, | |
| "grad_norm": 5.108152866363525, | |
| "learning_rate": 1.56e-06, | |
| "loss": 0.9862, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.04396678065461651, | |
| "grad_norm": 6.09797477722168, | |
| "learning_rate": 1.76e-06, | |
| "loss": 1.0003, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.048851978505129456, | |
| "grad_norm": 6.775981426239014, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.9876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0537371763556424, | |
| "grad_norm": 4.734734535217285, | |
| "learning_rate": 2.16e-06, | |
| "loss": 0.9026, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05862237420615535, | |
| "grad_norm": 5.119976043701172, | |
| "learning_rate": 2.3600000000000003e-06, | |
| "loss": 0.9783, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0635075720566683, | |
| "grad_norm": 4.984339714050293, | |
| "learning_rate": 2.56e-06, | |
| "loss": 0.9671, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.06839276990718124, | |
| "grad_norm": 5.38236665725708, | |
| "learning_rate": 2.7600000000000003e-06, | |
| "loss": 0.9263, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.07327796775769418, | |
| "grad_norm": 4.587165355682373, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.9063, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07816316560820713, | |
| "grad_norm": 4.583350658416748, | |
| "learning_rate": 3.1600000000000002e-06, | |
| "loss": 0.84, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.08304836345872008, | |
| "grad_norm": 4.574127674102783, | |
| "learning_rate": 3.3600000000000004e-06, | |
| "loss": 0.833, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.08793356130923302, | |
| "grad_norm": 4.4891486167907715, | |
| "learning_rate": 3.5600000000000002e-06, | |
| "loss": 0.7965, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.09281875915974597, | |
| "grad_norm": 4.560169696807861, | |
| "learning_rate": 3.7600000000000004e-06, | |
| "loss": 0.7703, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.09770395701025891, | |
| "grad_norm": 5.059728145599365, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.8702, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09770395701025891, | |
| "eval_loss": 0.7939865589141846, | |
| "eval_runtime": 4279.5784, | |
| "eval_samples_per_second": 0.956, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 55.69627284341499, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10258915486077186, | |
| "grad_norm": 4.930312156677246, | |
| "learning_rate": 4.16e-06, | |
| "loss": 0.7978, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.1074743527112848, | |
| "grad_norm": 4.10357666015625, | |
| "learning_rate": 4.360000000000001e-06, | |
| "loss": 0.7512, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.11235955056179775, | |
| "grad_norm": 4.250857830047607, | |
| "learning_rate": 4.56e-06, | |
| "loss": 0.7242, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.1172447484123107, | |
| "grad_norm": 4.367275714874268, | |
| "learning_rate": 4.76e-06, | |
| "loss": 0.7473, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.12212994626282364, | |
| "grad_norm": 5.22398042678833, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.8216, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1270151441133366, | |
| "grad_norm": 4.164461135864258, | |
| "learning_rate": 5.1600000000000006e-06, | |
| "loss": 0.6435, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.13190034196384953, | |
| "grad_norm": 5.729864120483398, | |
| "learning_rate": 5.36e-06, | |
| "loss": 0.7428, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.13678553981436248, | |
| "grad_norm": 4.261190891265869, | |
| "learning_rate": 5.560000000000001e-06, | |
| "loss": 0.7081, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.14167073766487542, | |
| "grad_norm": 3.940690279006958, | |
| "learning_rate": 5.76e-06, | |
| "loss": 0.63, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.14655593551538837, | |
| "grad_norm": 3.8406989574432373, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.6711, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.15144113336590131, | |
| "grad_norm": 3.9562718868255615, | |
| "learning_rate": 6.16e-06, | |
| "loss": 0.6992, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.15632633121641426, | |
| "grad_norm": 4.3271589279174805, | |
| "learning_rate": 6.360000000000001e-06, | |
| "loss": 0.7054, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.1612115290669272, | |
| "grad_norm": 4.195059776306152, | |
| "learning_rate": 6.560000000000001e-06, | |
| "loss": 0.6424, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.16609672691744015, | |
| "grad_norm": 4.491285800933838, | |
| "learning_rate": 6.760000000000001e-06, | |
| "loss": 0.7025, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.1709819247679531, | |
| "grad_norm": 5.474099636077881, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.6373, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.17586712261846604, | |
| "grad_norm": 4.050487995147705, | |
| "learning_rate": 7.16e-06, | |
| "loss": 0.5873, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.180752320468979, | |
| "grad_norm": 5.067830562591553, | |
| "learning_rate": 7.360000000000001e-06, | |
| "loss": 0.6215, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.18563751831949193, | |
| "grad_norm": 4.530464172363281, | |
| "learning_rate": 7.5600000000000005e-06, | |
| "loss": 0.6653, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.19052271617000488, | |
| "grad_norm": 3.94427490234375, | |
| "learning_rate": 7.76e-06, | |
| "loss": 0.6384, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.19540791402051783, | |
| "grad_norm": 4.769630432128906, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.6221, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19540791402051783, | |
| "eval_loss": 0.6273319721221924, | |
| "eval_runtime": 4292.0244, | |
| "eval_samples_per_second": 0.954, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 46.51759428984058, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.20029311187103077, | |
| "grad_norm": 5.084413528442383, | |
| "learning_rate": 8.16e-06, | |
| "loss": 0.6848, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.20517830972154372, | |
| "grad_norm": 4.525864124298096, | |
| "learning_rate": 8.36e-06, | |
| "loss": 0.6234, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.21006350757205666, | |
| "grad_norm": 4.084338188171387, | |
| "learning_rate": 8.560000000000001e-06, | |
| "loss": 0.6272, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.2149487054225696, | |
| "grad_norm": 4.2073187828063965, | |
| "learning_rate": 8.76e-06, | |
| "loss": 0.6075, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.21983390327308255, | |
| "grad_norm": 3.7465310096740723, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.5911, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.2247191011235955, | |
| "grad_norm": 3.5266239643096924, | |
| "learning_rate": 9.16e-06, | |
| "loss": 0.5499, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.22960429897410845, | |
| "grad_norm": 4.592888832092285, | |
| "learning_rate": 9.360000000000002e-06, | |
| "loss": 0.6124, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.2344894968246214, | |
| "grad_norm": 4.9025726318359375, | |
| "learning_rate": 9.56e-06, | |
| "loss": 0.5697, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.23937469467513434, | |
| "grad_norm": 4.163031101226807, | |
| "learning_rate": 9.760000000000001e-06, | |
| "loss": 0.6106, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.24425989252564728, | |
| "grad_norm": 5.005125045776367, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.6138, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24914509037616023, | |
| "grad_norm": 3.767075538635254, | |
| "learning_rate": 9.982222222222224e-06, | |
| "loss": 0.601, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2540302882266732, | |
| "grad_norm": 4.501799583435059, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.5664, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2589154860771861, | |
| "grad_norm": 4.360989570617676, | |
| "learning_rate": 9.937777777777779e-06, | |
| "loss": 0.5782, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.26380068392769906, | |
| "grad_norm": 3.4221303462982178, | |
| "learning_rate": 9.915555555555556e-06, | |
| "loss": 0.585, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.268685881778212, | |
| "grad_norm": 4.4359540939331055, | |
| "learning_rate": 9.893333333333334e-06, | |
| "loss": 0.6203, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.27357107962872496, | |
| "grad_norm": 4.127051830291748, | |
| "learning_rate": 9.871111111111112e-06, | |
| "loss": 0.5032, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.2784562774792379, | |
| "grad_norm": 4.420245170593262, | |
| "learning_rate": 9.84888888888889e-06, | |
| "loss": 0.5203, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.28334147532975085, | |
| "grad_norm": 3.9797379970550537, | |
| "learning_rate": 9.826666666666667e-06, | |
| "loss": 0.5655, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.2882266731802638, | |
| "grad_norm": 4.368413925170898, | |
| "learning_rate": 9.804444444444444e-06, | |
| "loss": 0.5212, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.29311187103077674, | |
| "grad_norm": 4.136660099029541, | |
| "learning_rate": 9.782222222222222e-06, | |
| "loss": 0.5569, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.29311187103077674, | |
| "eval_loss": 0.5524213910102844, | |
| "eval_runtime": 4286.0796, | |
| "eval_samples_per_second": 0.955, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 43.251471976892745, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2979970688812897, | |
| "grad_norm": 4.450418472290039, | |
| "learning_rate": 9.760000000000001e-06, | |
| "loss": 0.5764, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.30288226673180263, | |
| "grad_norm": 3.492920160293579, | |
| "learning_rate": 9.737777777777779e-06, | |
| "loss": 0.5373, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.3077674645823156, | |
| "grad_norm": 4.476955890655518, | |
| "learning_rate": 9.715555555555557e-06, | |
| "loss": 0.4991, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.3126526624328285, | |
| "grad_norm": 3.3835787773132324, | |
| "learning_rate": 9.693333333333334e-06, | |
| "loss": 0.5387, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.31753786028334147, | |
| "grad_norm": 3.3086416721343994, | |
| "learning_rate": 9.671111111111112e-06, | |
| "loss": 0.5414, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.3224230581338544, | |
| "grad_norm": 4.225744724273682, | |
| "learning_rate": 9.64888888888889e-06, | |
| "loss": 0.466, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.32730825598436736, | |
| "grad_norm": 4.349085807800293, | |
| "learning_rate": 9.626666666666667e-06, | |
| "loss": 0.5382, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.3321934538348803, | |
| "grad_norm": 3.8492863178253174, | |
| "learning_rate": 9.604444444444445e-06, | |
| "loss": 0.4948, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.33707865168539325, | |
| "grad_norm": 3.621649980545044, | |
| "learning_rate": 9.582222222222222e-06, | |
| "loss": 0.4967, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.3419638495359062, | |
| "grad_norm": 4.60444974899292, | |
| "learning_rate": 9.56e-06, | |
| "loss": 0.4764, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.34684904738641914, | |
| "grad_norm": 3.711239814758301, | |
| "learning_rate": 9.537777777777778e-06, | |
| "loss": 0.5223, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.3517342452369321, | |
| "grad_norm": 3.4989352226257324, | |
| "learning_rate": 9.515555555555557e-06, | |
| "loss": 0.5236, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.35661944308744503, | |
| "grad_norm": 3.1489815711975098, | |
| "learning_rate": 9.493333333333334e-06, | |
| "loss": 0.5124, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.361504640937958, | |
| "grad_norm": 4.3361735343933105, | |
| "learning_rate": 9.471111111111112e-06, | |
| "loss": 0.5418, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.3663898387884709, | |
| "grad_norm": 2.7804179191589355, | |
| "learning_rate": 9.44888888888889e-06, | |
| "loss": 0.4497, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.37127503663898387, | |
| "grad_norm": 3.681450605392456, | |
| "learning_rate": 9.426666666666667e-06, | |
| "loss": 0.4977, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.3761602344894968, | |
| "grad_norm": 4.99979829788208, | |
| "learning_rate": 9.404444444444445e-06, | |
| "loss": 0.5545, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.38104543234000976, | |
| "grad_norm": 3.4073972702026367, | |
| "learning_rate": 9.382222222222223e-06, | |
| "loss": 0.4368, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.3859306301905227, | |
| "grad_norm": 3.558271646499634, | |
| "learning_rate": 9.360000000000002e-06, | |
| "loss": 0.4994, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.39081582804103565, | |
| "grad_norm": 3.8624231815338135, | |
| "learning_rate": 9.33777777777778e-06, | |
| "loss": 0.4785, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.39081582804103565, | |
| "eval_loss": 0.5109024047851562, | |
| "eval_runtime": 4283.004, | |
| "eval_samples_per_second": 0.956, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 40.329389546186746, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3957010258915486, | |
| "grad_norm": 5.084819793701172, | |
| "learning_rate": 9.315555555555557e-06, | |
| "loss": 0.5334, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.40058622374206154, | |
| "grad_norm": 4.143725872039795, | |
| "learning_rate": 9.293333333333335e-06, | |
| "loss": 0.4547, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.4054714215925745, | |
| "grad_norm": 4.0059614181518555, | |
| "learning_rate": 9.271111111111112e-06, | |
| "loss": 0.5135, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.41035661944308743, | |
| "grad_norm": 4.0563483238220215, | |
| "learning_rate": 9.24888888888889e-06, | |
| "loss": 0.4474, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.4152418172936004, | |
| "grad_norm": 3.5411062240600586, | |
| "learning_rate": 9.226666666666668e-06, | |
| "loss": 0.5204, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.4201270151441133, | |
| "grad_norm": 3.515293836593628, | |
| "learning_rate": 9.204444444444445e-06, | |
| "loss": 0.514, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.42501221299462627, | |
| "grad_norm": 3.598378896713257, | |
| "learning_rate": 9.182222222222223e-06, | |
| "loss": 0.4408, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.4298974108451392, | |
| "grad_norm": 3.1513373851776123, | |
| "learning_rate": 9.16e-06, | |
| "loss": 0.4575, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 5.0970377922058105, | |
| "learning_rate": 9.137777777777778e-06, | |
| "loss": 0.4901, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.4396678065461651, | |
| "grad_norm": 4.093622207641602, | |
| "learning_rate": 9.115555555555556e-06, | |
| "loss": 0.4744, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.44455300439667805, | |
| "grad_norm": 4.782330513000488, | |
| "learning_rate": 9.093333333333333e-06, | |
| "loss": 0.4592, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.449438202247191, | |
| "grad_norm": 2.7775630950927734, | |
| "learning_rate": 9.07111111111111e-06, | |
| "loss": 0.511, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.45432340009770394, | |
| "grad_norm": 3.0217251777648926, | |
| "learning_rate": 9.048888888888888e-06, | |
| "loss": 0.4973, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.4592085979482169, | |
| "grad_norm": 4.689064025878906, | |
| "learning_rate": 9.026666666666666e-06, | |
| "loss": 0.5676, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.46409379579872984, | |
| "grad_norm": 3.9128365516662598, | |
| "learning_rate": 9.004444444444445e-06, | |
| "loss": 0.4762, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.4689789936492428, | |
| "grad_norm": 2.82621169090271, | |
| "learning_rate": 8.982222222222223e-06, | |
| "loss": 0.4318, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.4738641914997557, | |
| "grad_norm": 3.4691481590270996, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.4465, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.4787493893502687, | |
| "grad_norm": 4.63604736328125, | |
| "learning_rate": 8.937777777777778e-06, | |
| "loss": 0.4895, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.4836345872007816, | |
| "grad_norm": 4.685360431671143, | |
| "learning_rate": 8.915555555555556e-06, | |
| "loss": 0.5155, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.48851978505129456, | |
| "grad_norm": 3.411714553833008, | |
| "learning_rate": 8.893333333333333e-06, | |
| "loss": 0.5216, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48851978505129456, | |
| "eval_loss": 0.48320430517196655, | |
| "eval_runtime": 4282.5384, | |
| "eval_samples_per_second": 0.956, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 38.85151085930122, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4934049829018075, | |
| "grad_norm": 3.4970908164978027, | |
| "learning_rate": 8.871111111111111e-06, | |
| "loss": 0.4371, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.49829018075232046, | |
| "grad_norm": 4.287569522857666, | |
| "learning_rate": 8.848888888888889e-06, | |
| "loss": 0.4882, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.5031753786028335, | |
| "grad_norm": 4.18141508102417, | |
| "learning_rate": 8.826666666666668e-06, | |
| "loss": 0.4955, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.5080605764533463, | |
| "grad_norm": 3.6604855060577393, | |
| "learning_rate": 8.804444444444446e-06, | |
| "loss": 0.4448, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.5129457743038593, | |
| "grad_norm": 4.121953010559082, | |
| "learning_rate": 8.782222222222223e-06, | |
| "loss": 0.5134, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.5178309721543722, | |
| "grad_norm": 4.355337142944336, | |
| "learning_rate": 8.76e-06, | |
| "loss": 0.5069, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.5227161700048852, | |
| "grad_norm": 3.3341116905212402, | |
| "learning_rate": 8.737777777777778e-06, | |
| "loss": 0.5008, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.5276013678553981, | |
| "grad_norm": 4.42513370513916, | |
| "learning_rate": 8.715555555555556e-06, | |
| "loss": 0.4557, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.5324865657059111, | |
| "grad_norm": 2.6197452545166016, | |
| "learning_rate": 8.693333333333334e-06, | |
| "loss": 0.458, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.537371763556424, | |
| "grad_norm": 2.7881979942321777, | |
| "learning_rate": 8.671111111111113e-06, | |
| "loss": 0.4692, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.542256961406937, | |
| "grad_norm": 3.419318914413452, | |
| "learning_rate": 8.64888888888889e-06, | |
| "loss": 0.4541, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.5471421592574499, | |
| "grad_norm": 3.8821029663085938, | |
| "learning_rate": 8.626666666666668e-06, | |
| "loss": 0.5263, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.5520273571079629, | |
| "grad_norm": 3.7360761165618896, | |
| "learning_rate": 8.604444444444446e-06, | |
| "loss": 0.479, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.5569125549584758, | |
| "grad_norm": 3.3444161415100098, | |
| "learning_rate": 8.582222222222223e-06, | |
| "loss": 0.4942, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.5617977528089888, | |
| "grad_norm": 3.6646621227264404, | |
| "learning_rate": 8.560000000000001e-06, | |
| "loss": 0.4129, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.5666829506595017, | |
| "grad_norm": 3.4366016387939453, | |
| "learning_rate": 8.537777777777779e-06, | |
| "loss": 0.4324, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.5715681485100147, | |
| "grad_norm": 3.347890615463257, | |
| "learning_rate": 8.515555555555556e-06, | |
| "loss": 0.4797, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.5764533463605276, | |
| "grad_norm": 3.3374993801116943, | |
| "learning_rate": 8.493333333333334e-06, | |
| "loss": 0.4665, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.5813385442110406, | |
| "grad_norm": 3.522141933441162, | |
| "learning_rate": 8.471111111111112e-06, | |
| "loss": 0.4141, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.5862237420615535, | |
| "grad_norm": 3.1512022018432617, | |
| "learning_rate": 8.448888888888889e-06, | |
| "loss": 0.3981, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5862237420615535, | |
| "eval_loss": 0.4652426540851593, | |
| "eval_runtime": 4278.5694, | |
| "eval_samples_per_second": 0.957, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 37.93360828750764, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.540839686144e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |