| { | |
| "best_metric": 0.19929151237010956, | |
| "best_model_checkpoint": "outputs/bart-base-detox/checkpoint-405", | |
| "epoch": 3.9953917050691246, | |
| "global_step": 540, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9e-06, | |
| "loss": 0.5633, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.25235414505004883, | |
| "eval_runtime": 12.2006, | |
| "eval_samples_per_second": 254.085, | |
| "eval_steps_per_second": 31.802, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2589, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.21927106380462646, | |
| "eval_runtime": 11.2792, | |
| "eval_samples_per_second": 274.842, | |
| "eval_steps_per_second": 34.4, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7e-06, | |
| "loss": 0.2307, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.19929151237010956, | |
| "eval_runtime": 11.2313, | |
| "eval_samples_per_second": 276.016, | |
| "eval_steps_per_second": 34.546, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.2171, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.20016591250896454, | |
| "eval_runtime": 11.0923, | |
| "eval_samples_per_second": 279.473, | |
| "eval_steps_per_second": 34.979, | |
| "step": 540 | |
| } | |
| ], | |
| "max_steps": 1350, | |
| "num_train_epochs": 10, | |
| "total_flos": 2151917158809600.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |