| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 3720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0002774585635359116, | |
| "loss": 4.2459, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.2878854274749756, | |
| "eval_runtime": 169.7192, | |
| "eval_samples_per_second": 10.865, | |
| "eval_wer": 1.0, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.00024662983425414363, | |
| "loss": 0.6602, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.39920976758003235, | |
| "eval_runtime": 174.6666, | |
| "eval_samples_per_second": 10.557, | |
| "eval_wer": 0.49533842998321836, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 0.00021580110497237566, | |
| "loss": 0.2499, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.362725168466568, | |
| "eval_runtime": 177.9959, | |
| "eval_samples_per_second": 10.36, | |
| "eval_wer": 0.4402386723848592, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 0.00018497237569060774, | |
| "loss": 0.1844, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.3489285111427307, | |
| "eval_runtime": 183.1767, | |
| "eval_samples_per_second": 10.067, | |
| "eval_wer": 0.4130151034868544, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.00015414364640883976, | |
| "loss": 0.1414, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.3539421558380127, | |
| "eval_runtime": 179.9374, | |
| "eval_samples_per_second": 10.248, | |
| "eval_wer": 0.3987506992355025, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 0.00012331491712707182, | |
| "loss": 0.1176, | |
| "step": 2232 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 0.38211411237716675, | |
| "eval_runtime": 183.8348, | |
| "eval_samples_per_second": 10.031, | |
| "eval_wer": 0.39800484803281744, | |
| "step": 2232 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 9.248618784530387e-05, | |
| "loss": 0.1014, | |
| "step": 2604 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 0.3974383771419525, | |
| "eval_runtime": 184.5532, | |
| "eval_samples_per_second": 9.992, | |
| "eval_wer": 0.39203803841133694, | |
| "step": 2604 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 6.165745856353591e-05, | |
| "loss": 0.0884, | |
| "step": 2976 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.3847784698009491, | |
| "eval_runtime": 186.9619, | |
| "eval_samples_per_second": 9.863, | |
| "eval_wer": 0.38868170799925417, | |
| "step": 2976 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "learning_rate": 3.0828729281767954e-05, | |
| "loss": 0.0806, | |
| "step": 3348 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_loss": 0.38000500202178955, | |
| "eval_runtime": 190.0136, | |
| "eval_samples_per_second": 9.705, | |
| "eval_wer": 0.3840201379824725, | |
| "step": 3348 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0748, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 0.38302138447761536, | |
| "eval_runtime": 191.9169, | |
| "eval_samples_per_second": 9.608, | |
| "eval_wer": 0.3817825843744173, | |
| "step": 3720 | |
| } | |
| ], | |
| "max_steps": 3720, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.797456336593924e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |