| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.733485193621868, | |
| "eval_steps": 200, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.45558086560364464, | |
| "eval_loss": 0.6622544527053833, | |
| "eval_runtime": 198.7007, | |
| "eval_samples_per_second": 35.34, | |
| "eval_steps_per_second": 0.554, | |
| "eval_wer": 0.49498398862594506, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "eval_loss": 0.5531215667724609, | |
| "eval_runtime": 198.3629, | |
| "eval_samples_per_second": 35.4, | |
| "eval_steps_per_second": 0.555, | |
| "eval_wer": 0.41153881927742864, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 0.38089731335639954, | |
| "learning_rate": 0.0001969444444444444, | |
| "loss": 1.5958, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "eval_loss": 0.5106531381607056, | |
| "eval_runtime": 200.3157, | |
| "eval_samples_per_second": 35.055, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 0.38682416720923746, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8223234624145785, | |
| "eval_loss": 0.5462459325790405, | |
| "eval_runtime": 199.4037, | |
| "eval_samples_per_second": 35.215, | |
| "eval_steps_per_second": 0.552, | |
| "eval_wer": 0.399859153058106, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.277904328018223, | |
| "grad_norm": 2.8862736225128174, | |
| "learning_rate": 5.972222222222221e-05, | |
| "loss": 0.6374, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.277904328018223, | |
| "eval_loss": 0.5843283534049988, | |
| "eval_runtime": 200.2605, | |
| "eval_samples_per_second": 35.064, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 0.38792702533916207, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.733485193621868, | |
| "eval_loss": 0.6271677017211914, | |
| "eval_runtime": 200.7317, | |
| "eval_samples_per_second": 34.982, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.38699690402476783, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.733485193621868, | |
| "step": 1200, | |
| "total_flos": 9.105570625839254e+18, | |
| "train_loss": 1.0454113133748373, | |
| "train_runtime": 3514.2413, | |
| "train_samples_per_second": 21.854, | |
| "train_steps_per_second": 0.341 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1200, | |
| "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": 9.105570625839254e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |