| { | |
| "best_metric": 27.250397341424648, | |
| "best_model_checkpoint": "./checkpoint-50", | |
| "epoch": 99.01, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 9.210526315789473e-07, | |
| "loss": 0.752, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 7.947368421052631e-07, | |
| "loss": 0.425, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "eval_loss": 0.4105507731437683, | |
| "eval_runtime": 786.9933, | |
| "eval_samples_per_second": 1.006, | |
| "eval_steps_per_second": 0.032, | |
| "eval_wer": 27.250397341424648, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 6.63157894736842e-07, | |
| "loss": 0.2879, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 5.31578947368421e-07, | |
| "loss": 0.1906, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "eval_loss": 0.44195765256881714, | |
| "eval_runtime": 813.6355, | |
| "eval_samples_per_second": 0.973, | |
| "eval_steps_per_second": 0.031, | |
| "eval_wer": 29.013148388961135, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 4e-07, | |
| "loss": 0.1347, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 74.01, | |
| "learning_rate": 2.684210526315789e-07, | |
| "loss": 0.0982, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 74.01, | |
| "eval_loss": 0.4795360267162323, | |
| "eval_runtime": 848.8397, | |
| "eval_samples_per_second": 0.933, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 30.306314116457163, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "learning_rate": 1.368421052631579e-07, | |
| "loss": 0.0829, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 99.01, | |
| "learning_rate": 5.263157894736842e-09, | |
| "loss": 0.0717, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 99.01, | |
| "eval_loss": 0.4944527745246887, | |
| "eval_runtime": 875.6338, | |
| "eval_samples_per_second": 0.904, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 30.891489669122958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 99.01, | |
| "step": 200, | |
| "total_flos": 1.387640884064256e+20, | |
| "train_loss": 0.25535590410232545, | |
| "train_runtime": 48457.1181, | |
| "train_samples_per_second": 2.113, | |
| "train_steps_per_second": 0.004 | |
| } | |
| ], | |
| "max_steps": 200, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.387640884064256e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |