| { | |
| "best_metric": 0.32531991600990295, | |
| "best_model_checkpoint": "outputs/electra-nli-efl/checkpoint-630", | |
| "epoch": 2.9955555555555557, | |
| "global_step": 630, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.499999999999999e-07, | |
| "loss": 0.4175, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7316666666666667, | |
| "eval_f1": 0.7305281480923569, | |
| "eval_loss": 0.4019607603549957, | |
| "eval_runtime": 11.5986, | |
| "eval_samples_per_second": 258.651, | |
| "eval_steps_per_second": 32.331, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9e-07, | |
| "loss": 0.3061, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.768, | |
| "eval_f1": 0.7674936528439713, | |
| "eval_loss": 0.3519645631313324, | |
| "eval_runtime": 11.6033, | |
| "eval_samples_per_second": 258.548, | |
| "eval_steps_per_second": 32.319, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.499999999999999e-07, | |
| "loss": 0.2588, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.79, | |
| "eval_f1": 0.7887799932409598, | |
| "eval_loss": 0.32531991600990295, | |
| "eval_runtime": 11.5951, | |
| "eval_samples_per_second": 258.729, | |
| "eval_steps_per_second": 32.341, | |
| "step": 630 | |
| } | |
| ], | |
| "max_steps": 4200, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.2470681674040832e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |