| { | |
| "best_metric": 0.3247086703777313, | |
| "best_model_checkpoint": "outputs/electra-nli-efl/checkpoint-1050", | |
| "epoch": 5.995555555555556, | |
| "global_step": 1260, | |
| "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 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8e-07, | |
| "loss": 0.234, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.788, | |
| "eval_f1": 0.7877041067016072, | |
| "eval_loss": 0.33730262517929077, | |
| "eval_runtime": 11.6935, | |
| "eval_samples_per_second": 256.553, | |
| "eval_steps_per_second": 32.069, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.2116, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.804, | |
| "eval_f1": 0.8032759681298146, | |
| "eval_loss": 0.3247086703777313, | |
| "eval_runtime": 11.5938, | |
| "eval_samples_per_second": 258.758, | |
| "eval_steps_per_second": 32.345, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7e-07, | |
| "loss": 0.1974, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.793, | |
| "eval_f1": 0.7928414316247181, | |
| "eval_loss": 0.34000951051712036, | |
| "eval_runtime": 11.6038, | |
| "eval_samples_per_second": 258.537, | |
| "eval_steps_per_second": 32.317, | |
| "step": 1260 | |
| } | |
| ], | |
| "max_steps": 4200, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.494171282352371e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |