| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.96969696969697, | |
| "global_step": 960, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 3.88e-06, | |
| "loss": 8.6025, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 7.88e-06, | |
| "loss": 2.9845, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.73, | |
| "learning_rate": 1.188e-05, | |
| "loss": 2.8172, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 24.97, | |
| "learning_rate": 1.588e-05, | |
| "loss": 2.1658, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 31.24, | |
| "learning_rate": 1.9880000000000003e-05, | |
| "loss": 1.7135, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 31.24, | |
| "eval_loss": 0.9609270691871643, | |
| "eval_runtime": 24.5357, | |
| "eval_samples_per_second": 20.745, | |
| "eval_steps_per_second": 1.304, | |
| "eval_wer": 0.692578468759167, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 37.48, | |
| "learning_rate": 1.5782608695652177e-05, | |
| "loss": 1.4495, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 43.73, | |
| "learning_rate": 1.1434782608695654e-05, | |
| "loss": 1.253, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 49.97, | |
| "learning_rate": 7.086956521739131e-06, | |
| "loss": 1.1265, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 56.24, | |
| "learning_rate": 2.7391304347826087e-06, | |
| "loss": 1.0423, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 59.97, | |
| "step": 960, | |
| "total_flos": 2.336891900871078e+19, | |
| "train_loss": 2.4739840825398765, | |
| "train_runtime": 4674.7861, | |
| "train_samples_per_second": 13.425, | |
| "train_steps_per_second": 0.205 | |
| } | |
| ], | |
| "max_steps": 960, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.336891900871078e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |