| { | |
| "best_global_step": 800, | |
| "best_metric": 0.0701018573996405, | |
| "best_model_checkpoint": "/content/drive/MyDrive/ABA Projects/Speech-To-Text/models/Under9/800KB_aug_pitch/checkpoint-800", | |
| "epoch": 3.0, | |
| "eval_steps": 200, | |
| "global_step": 855, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 0.48045819997787476, | |
| "learning_rate": 3.8e-07, | |
| "loss": 0.0065, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 1.012063980102539, | |
| "learning_rate": 7.8e-07, | |
| "loss": 0.0074, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 0.4023517072200775, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.0067, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 0.5677161812782288, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 0.0057, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 0.6357490420341492, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.0063, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 0.5882821083068848, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.0067, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 0.4974755048751831, | |
| "learning_rate": 2.7800000000000005e-06, | |
| "loss": 0.0079, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 0.6867635846138, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.0064, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 0.40168455243110657, | |
| "learning_rate": 3.58e-06, | |
| "loss": 0.0075, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 0.7916053533554077, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 0.0067, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "eval_loss": 0.017162702977657318, | |
| "eval_runtime": 284.7635, | |
| "eval_samples_per_second": 1.686, | |
| "eval_steps_per_second": 0.053, | |
| "eval_wer": 0.08732774116237269, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "grad_norm": 0.5882860422134399, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.008, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 0.5530531406402588, | |
| "learning_rate": 4.78e-06, | |
| "loss": 0.0069, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "grad_norm": 0.4040575325489044, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.0082, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "grad_norm": 0.8813060522079468, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 0.0091, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 0.3994963467121124, | |
| "learning_rate": 5.98e-06, | |
| "loss": 0.0058, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1228070175438596, | |
| "grad_norm": 0.4101634621620178, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 0.0082, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.1929824561403508, | |
| "grad_norm": 0.6057800054550171, | |
| "learning_rate": 6.780000000000001e-06, | |
| "loss": 0.0074, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "grad_norm": 0.5515829920768738, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 0.006, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.6737426519393921, | |
| "learning_rate": 7.58e-06, | |
| "loss": 0.007, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 0.4856266379356384, | |
| "learning_rate": 7.980000000000002e-06, | |
| "loss": 0.0086, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "eval_loss": 0.020672131329774857, | |
| "eval_runtime": 284.5177, | |
| "eval_samples_per_second": 1.687, | |
| "eval_steps_per_second": 0.053, | |
| "eval_wer": 0.09781306171360096, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4736842105263157, | |
| "grad_norm": 0.771392285823822, | |
| "learning_rate": 8.380000000000001e-06, | |
| "loss": 0.0095, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.543859649122807, | |
| "grad_norm": 0.637648344039917, | |
| "learning_rate": 8.78e-06, | |
| "loss": 0.0078, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.6140350877192984, | |
| "grad_norm": 0.6620879769325256, | |
| "learning_rate": 9.180000000000002e-06, | |
| "loss": 0.0094, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.6842105263157894, | |
| "grad_norm": 1.040582537651062, | |
| "learning_rate": 9.58e-06, | |
| "loss": 0.0097, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 1.0017151832580566, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 0.01, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8245614035087718, | |
| "grad_norm": 0.4564901292324066, | |
| "learning_rate": 9.492957746478874e-06, | |
| "loss": 0.0118, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "grad_norm": 0.8653856515884399, | |
| "learning_rate": 8.929577464788734e-06, | |
| "loss": 0.0125, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.9649122807017543, | |
| "grad_norm": 1.0773769617080688, | |
| "learning_rate": 8.366197183098592e-06, | |
| "loss": 0.01, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.0350877192982457, | |
| "grad_norm": 0.6844319105148315, | |
| "learning_rate": 7.80281690140845e-06, | |
| "loss": 0.0093, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 1.1129411458969116, | |
| "learning_rate": 7.239436619718311e-06, | |
| "loss": 0.007, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "eval_loss": 0.021361183375120163, | |
| "eval_runtime": 284.3183, | |
| "eval_samples_per_second": 1.688, | |
| "eval_steps_per_second": 0.053, | |
| "eval_wer": 0.09032354703415219, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.175438596491228, | |
| "grad_norm": 0.5749935507774353, | |
| "learning_rate": 6.67605633802817e-06, | |
| "loss": 0.0064, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.245614035087719, | |
| "grad_norm": 0.6366358399391174, | |
| "learning_rate": 6.112676056338028e-06, | |
| "loss": 0.0059, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.3157894736842106, | |
| "grad_norm": 0.4254266917705536, | |
| "learning_rate": 5.549295774647888e-06, | |
| "loss": 0.0043, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.3859649122807016, | |
| "grad_norm": 0.7417527437210083, | |
| "learning_rate": 4.985915492957747e-06, | |
| "loss": 0.0041, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.456140350877193, | |
| "grad_norm": 0.32039281725883484, | |
| "learning_rate": 4.422535211267606e-06, | |
| "loss": 0.0037, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.526315789473684, | |
| "grad_norm": 0.5453156232833862, | |
| "learning_rate": 3.859154929577465e-06, | |
| "loss": 0.0028, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.5964912280701755, | |
| "grad_norm": 0.3248697519302368, | |
| "learning_rate": 3.295774647887324e-06, | |
| "loss": 0.0053, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.27672505378723145, | |
| "learning_rate": 2.7323943661971837e-06, | |
| "loss": 0.0035, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.736842105263158, | |
| "grad_norm": 0.2517031133174896, | |
| "learning_rate": 2.169014084507042e-06, | |
| "loss": 0.0036, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "grad_norm": 0.2080334722995758, | |
| "learning_rate": 1.6056338028169016e-06, | |
| "loss": 0.0038, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "eval_loss": 0.014348218217492104, | |
| "eval_runtime": 284.6078, | |
| "eval_samples_per_second": 1.687, | |
| "eval_steps_per_second": 0.053, | |
| "eval_wer": 0.0701018573996405, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.8771929824561404, | |
| "grad_norm": 0.38768911361694336, | |
| "learning_rate": 1.0422535211267606e-06, | |
| "loss": 0.0032, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.9473684210526314, | |
| "grad_norm": 0.40687140822410583, | |
| "learning_rate": 4.788732394366198e-07, | |
| "loss": 0.0027, | |
| "step": 840 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 855, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.661165956202496e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |