| { | |
| "best_metric": 0.04146179184317589, | |
| "best_model_checkpoint": "./models/deberta-v3-base-nli-v1/checkpoint-811", | |
| "epoch": 3.0, | |
| "global_step": 2433, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 5.56144677353062e-06, | |
| "loss": 0.064, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.04146179184317589, | |
| "eval_rmse": 0.20362168550491333, | |
| "eval_runtime": 4.397, | |
| "eval_samples_per_second": 348.191, | |
| "eval_steps_per_second": 29.111, | |
| "step": 811 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.122893547061241e-06, | |
| "loss": 0.0441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.684340320591862e-06, | |
| "loss": 0.0347, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.04254383593797684, | |
| "eval_rmse": 0.20626157522201538, | |
| "eval_runtime": 4.3965, | |
| "eval_samples_per_second": 348.229, | |
| "eval_steps_per_second": 29.114, | |
| "step": 1622 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.2457870941224826e-06, | |
| "loss": 0.0292, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.04311966523528099, | |
| "eval_rmse": 0.207652747631073, | |
| "eval_runtime": 4.4067, | |
| "eval_samples_per_second": 347.422, | |
| "eval_steps_per_second": 29.046, | |
| "step": 2433 | |
| } | |
| ], | |
| "max_steps": 2433, | |
| "num_train_epochs": 3, | |
| "total_flos": 1826338023730968.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |