| { | |
| "best_metric": 0.6086750030517578, | |
| "best_model_checkpoint": "hBERTv1_data_aug_sst2/checkpoint-4374", | |
| "epoch": 6.0, | |
| "global_step": 26244, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.2892, | |
| "step": 4374 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7855504587155964, | |
| "eval_loss": 0.6086750030517578, | |
| "eval_runtime": 1.1244, | |
| "eval_samples_per_second": 775.522, | |
| "eval_steps_per_second": 3.557, | |
| "step": 4374 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8000228623685414e-05, | |
| "loss": 0.1628, | |
| "step": 8748 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7809633027522935, | |
| "eval_loss": 0.7398399114608765, | |
| "eval_runtime": 1.1251, | |
| "eval_samples_per_second": 775.039, | |
| "eval_steps_per_second": 3.555, | |
| "step": 8748 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700068587105624e-05, | |
| "loss": 0.1151, | |
| "step": 13122 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.801605504587156, | |
| "eval_loss": 0.84917813539505, | |
| "eval_runtime": 1.1267, | |
| "eval_samples_per_second": 773.943, | |
| "eval_steps_per_second": 3.55, | |
| "step": 13122 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600114311842707e-05, | |
| "loss": 0.0917, | |
| "step": 17496 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.786697247706422, | |
| "eval_loss": 1.0381159782409668, | |
| "eval_runtime": 1.1306, | |
| "eval_samples_per_second": 771.304, | |
| "eval_steps_per_second": 3.538, | |
| "step": 17496 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.500137174211249e-05, | |
| "loss": 0.0862, | |
| "step": 21870 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.786697247706422, | |
| "eval_loss": 0.9656515121459961, | |
| "eval_runtime": 1.1426, | |
| "eval_samples_per_second": 763.183, | |
| "eval_steps_per_second": 3.501, | |
| "step": 21870 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400205761316873e-05, | |
| "loss": 0.0762, | |
| "step": 26244 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7821100917431193, | |
| "eval_loss": 1.0815026760101318, | |
| "eval_runtime": 1.1296, | |
| "eval_samples_per_second": 771.958, | |
| "eval_steps_per_second": 3.541, | |
| "step": 26244 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 26244, | |
| "total_flos": 8.48521401219416e+17, | |
| "train_loss": 0.1368842709992939, | |
| "train_runtime": 24327.6462, | |
| "train_samples_per_second": 2301.125, | |
| "train_steps_per_second": 8.99 | |
| } | |
| ], | |
| "max_steps": 218700, | |
| "num_train_epochs": 50, | |
| "total_flos": 8.48521401219416e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |