| { | |
| "best_metric": 0.8213908076286316, | |
| "best_model_checkpoint": "./wav2vec2-base-960h-paper/checkpoint-3352", | |
| "epoch": 49.64200477326969, | |
| "global_step": 5200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 5.683954238891602, | |
| "eval_runtime": 61.9414, | |
| "eval_samples_per_second": 13.594, | |
| "eval_steps_per_second": 0.856, | |
| "eval_wer": 1.0, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.9771974086761475, | |
| "eval_runtime": 58.5656, | |
| "eval_samples_per_second": 14.377, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 1.0, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.4203968048095703, | |
| "eval_runtime": 58.5182, | |
| "eval_samples_per_second": 14.389, | |
| "eval_steps_per_second": 0.906, | |
| "eval_wer": 1.0, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.3691885471343994, | |
| "eval_runtime": 58.4394, | |
| "eval_samples_per_second": 14.408, | |
| "eval_steps_per_second": 0.907, | |
| "eval_wer": 1.0, | |
| "step": 419 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 4.916317991631799e-05, | |
| "loss": 5.5612, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_loss": 3.3944718837738037, | |
| "eval_runtime": 58.5781, | |
| "eval_samples_per_second": 14.374, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 1.0, | |
| "step": 523 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.342637777328491, | |
| "eval_runtime": 58.3305, | |
| "eval_samples_per_second": 14.435, | |
| "eval_steps_per_second": 0.909, | |
| "eval_wer": 1.0, | |
| "step": 628 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 3.3333232402801514, | |
| "eval_runtime": 58.4524, | |
| "eval_samples_per_second": 14.405, | |
| "eval_steps_per_second": 0.907, | |
| "eval_wer": 1.0, | |
| "step": 733 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.329634428024292, | |
| "eval_runtime": 58.6528, | |
| "eval_samples_per_second": 14.356, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 1.0000659917906212, | |
| "step": 838 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_loss": 3.1853039264678955, | |
| "eval_runtime": 58.7103, | |
| "eval_samples_per_second": 14.342, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9999340082093787, | |
| "step": 942 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 4.393305439330544e-05, | |
| "loss": 3.2743, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.1381497383117676, | |
| "eval_runtime": 58.7525, | |
| "eval_samples_per_second": 14.331, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 1.0245489461111037, | |
| "step": 1047 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.6965396404266357, | |
| "eval_runtime": 58.7454, | |
| "eval_samples_per_second": 14.333, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 1.0141750366254438, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.4229621887207031, | |
| "eval_runtime": 58.8247, | |
| "eval_samples_per_second": 14.314, | |
| "eval_steps_per_second": 0.901, | |
| "eval_wer": 1.0011350587986854, | |
| "step": 1257 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_loss": 1.2678765058517456, | |
| "eval_runtime": 58.7761, | |
| "eval_samples_per_second": 14.326, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9873295762007206, | |
| "step": 1361 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.1569700241088867, | |
| "eval_runtime": 58.6837, | |
| "eval_samples_per_second": 14.348, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9835944408515581, | |
| "step": 1466 | |
| }, | |
| { | |
| "epoch": 14.32, | |
| "learning_rate": 3.870292887029289e-05, | |
| "loss": 1.5432, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.0858139991760254, | |
| "eval_runtime": 58.71, | |
| "eval_samples_per_second": 14.342, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9784206844668524, | |
| "step": 1571 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.0303007364273071, | |
| "eval_runtime": 58.7899, | |
| "eval_samples_per_second": 14.322, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9768632782081909, | |
| "step": 1676 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_loss": 0.985497236251831, | |
| "eval_runtime": 58.9329, | |
| "eval_samples_per_second": 14.287, | |
| "eval_steps_per_second": 0.899, | |
| "eval_wer": 0.97459316061082, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.9558756947517395, | |
| "eval_runtime": 58.7163, | |
| "eval_samples_per_second": 14.34, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9709240170522787, | |
| "step": 1885 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.93276047706604, | |
| "eval_runtime": 58.6741, | |
| "eval_samples_per_second": 14.35, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9727981839059221, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 3.3472803347280334e-05, | |
| "loss": 0.902, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.916550874710083, | |
| "eval_runtime": 58.8598, | |
| "eval_samples_per_second": 14.305, | |
| "eval_steps_per_second": 0.9, | |
| "eval_wer": 0.9738144574814893, | |
| "step": 2095 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_loss": 0.8990672826766968, | |
| "eval_runtime": 58.8033, | |
| "eval_samples_per_second": 14.319, | |
| "eval_steps_per_second": 0.901, | |
| "eval_wer": 0.9697625615373447, | |
| "step": 2199 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.8716745376586914, | |
| "eval_runtime": 58.795, | |
| "eval_samples_per_second": 14.321, | |
| "eval_steps_per_second": 0.901, | |
| "eval_wer": 0.9680731716974409, | |
| "step": 2304 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.8665304780006409, | |
| "eval_runtime": 58.9314, | |
| "eval_samples_per_second": 14.288, | |
| "eval_steps_per_second": 0.899, | |
| "eval_wer": 0.9669249145406311, | |
| "step": 2409 | |
| }, | |
| { | |
| "epoch": 23.87, | |
| "learning_rate": 2.824267782426778e-05, | |
| "loss": 0.7003, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.8588829636573792, | |
| "eval_runtime": 58.6586, | |
| "eval_samples_per_second": 14.354, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9670173030475009, | |
| "step": 2514 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_loss": 0.8419524431228638, | |
| "eval_runtime": 58.7876, | |
| "eval_samples_per_second": 14.323, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9659086409650639, | |
| "step": 2618 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.8473154306411743, | |
| "eval_runtime": 59.0243, | |
| "eval_samples_per_second": 14.265, | |
| "eval_steps_per_second": 0.898, | |
| "eval_wer": 0.9660538229044306, | |
| "step": 2723 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.8542926907539368, | |
| "eval_runtime": 58.7124, | |
| "eval_samples_per_second": 14.341, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9666345506618976, | |
| "step": 2828 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.8315206170082092, | |
| "eval_runtime": 58.7261, | |
| "eval_samples_per_second": 14.338, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9622922908390196, | |
| "step": 2933 | |
| }, | |
| { | |
| "epoch": 28.64, | |
| "learning_rate": 2.301255230125523e-05, | |
| "loss": 0.5914, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_loss": 0.8280870914459229, | |
| "eval_runtime": 58.9202, | |
| "eval_samples_per_second": 14.291, | |
| "eval_steps_per_second": 0.9, | |
| "eval_wer": 0.9625694563596289, | |
| "step": 3037 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.8315181732177734, | |
| "eval_runtime": 58.6719, | |
| "eval_samples_per_second": 14.351, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9625430596433804, | |
| "step": 3142 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 0.826146125793457, | |
| "eval_runtime": 58.8509, | |
| "eval_samples_per_second": 14.307, | |
| "eval_steps_per_second": 0.901, | |
| "eval_wer": 0.9619623318859134, | |
| "step": 3247 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.8213908076286316, | |
| "eval_runtime": 58.6666, | |
| "eval_samples_per_second": 14.352, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9639684823207993, | |
| "step": 3352 | |
| }, | |
| { | |
| "epoch": 32.99, | |
| "eval_loss": 0.8309536576271057, | |
| "eval_runtime": 58.499, | |
| "eval_samples_per_second": 14.393, | |
| "eval_steps_per_second": 0.906, | |
| "eval_wer": 0.9633745562052081, | |
| "step": 3456 | |
| }, | |
| { | |
| "epoch": 33.41, | |
| "learning_rate": 1.7782426778242676e-05, | |
| "loss": 0.5157, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.8252176642417908, | |
| "eval_runtime": 58.5446, | |
| "eval_samples_per_second": 14.382, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 0.9634669447120778, | |
| "step": 3561 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 0.8372853398323059, | |
| "eval_runtime": 58.5403, | |
| "eval_samples_per_second": 14.383, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 0.9638101020233083, | |
| "step": 3666 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.8421617746353149, | |
| "eval_runtime": 58.6533, | |
| "eval_samples_per_second": 14.356, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9629258120289836, | |
| "step": 3771 | |
| }, | |
| { | |
| "epoch": 36.99, | |
| "eval_loss": 0.8294224143028259, | |
| "eval_runtime": 58.6527, | |
| "eval_samples_per_second": 14.356, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9631897791914685, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.8332000374794006, | |
| "eval_runtime": 58.5856, | |
| "eval_samples_per_second": 14.372, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 0.9576332704211596, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 38.19, | |
| "learning_rate": 1.2552301255230125e-05, | |
| "loss": 0.4655, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 0.8329895734786987, | |
| "eval_runtime": 58.7603, | |
| "eval_samples_per_second": 14.329, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.959507437274803, | |
| "step": 4085 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.8296785950660706, | |
| "eval_runtime": 58.6153, | |
| "eval_samples_per_second": 14.365, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9625034645690076, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 40.99, | |
| "eval_loss": 0.8365075588226318, | |
| "eval_runtime": 58.6008, | |
| "eval_samples_per_second": 14.368, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9620547203927832, | |
| "step": 4294 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 0.8360996842384338, | |
| "eval_runtime": 58.6605, | |
| "eval_samples_per_second": 14.354, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9621339105415286, | |
| "step": 4399 | |
| }, | |
| { | |
| "epoch": 42.96, | |
| "learning_rate": 7.3221757322175736e-06, | |
| "loss": 0.4266, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 0.8416324853897095, | |
| "eval_runtime": 58.5672, | |
| "eval_samples_per_second": 14.377, | |
| "eval_steps_per_second": 0.905, | |
| "eval_wer": 0.9625034645690076, | |
| "step": 4504 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 0.8381497859954834, | |
| "eval_runtime": 58.6851, | |
| "eval_samples_per_second": 14.348, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9633745562052081, | |
| "step": 4609 | |
| }, | |
| { | |
| "epoch": 44.99, | |
| "eval_loss": 0.8447730541229248, | |
| "eval_runtime": 58.6902, | |
| "eval_samples_per_second": 14.347, | |
| "eval_steps_per_second": 0.903, | |
| "eval_wer": 0.9644964166457692, | |
| "step": 4713 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 0.8447101712226868, | |
| "eval_runtime": 58.8118, | |
| "eval_samples_per_second": 14.317, | |
| "eval_steps_per_second": 0.901, | |
| "eval_wer": 0.9624902662108834, | |
| "step": 4818 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 0.8463994264602661, | |
| "eval_runtime": 58.6459, | |
| "eval_samples_per_second": 14.357, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 0.9641268626182903, | |
| "step": 4923 | |
| }, | |
| { | |
| "epoch": 47.73, | |
| "learning_rate": 2.092050209205021e-06, | |
| "loss": 0.4019, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.8449317216873169, | |
| "eval_runtime": 58.893, | |
| "eval_samples_per_second": 14.297, | |
| "eval_steps_per_second": 0.9, | |
| "eval_wer": 0.9627542333733684, | |
| "step": 5028 | |
| }, | |
| { | |
| "epoch": 48.99, | |
| "eval_loss": 0.8487057089805603, | |
| "eval_runtime": 58.7287, | |
| "eval_samples_per_second": 14.337, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9625562580015046, | |
| "step": 5132 | |
| }, | |
| { | |
| "epoch": 49.64, | |
| "eval_loss": 0.846527636051178, | |
| "eval_runtime": 58.7826, | |
| "eval_samples_per_second": 14.324, | |
| "eval_steps_per_second": 0.902, | |
| "eval_wer": 0.9629126136708593, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 49.64, | |
| "step": 5200, | |
| "total_flos": 4.180377449644987e+19, | |
| "train_loss": 1.3979950288625864, | |
| "train_runtime": 89308.2068, | |
| "train_samples_per_second": 1.876, | |
| "train_steps_per_second": 0.058 | |
| } | |
| ], | |
| "max_steps": 5200, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.180377449644987e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |