| { | |
| "best_metric": 0.7581227436823105, | |
| "best_model_checkpoint": "./fine-tune/roberta-base/rte/checkpoint-780", | |
| "epoch": 8.0, | |
| "global_step": 1248, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7072076797485352, | |
| "eval_runtime": 0.4902, | |
| "eval_samples_per_second": 565.102, | |
| "eval_steps_per_second": 71.403, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6958328485488892, | |
| "eval_runtime": 0.689, | |
| "eval_samples_per_second": 402.039, | |
| "eval_steps_per_second": 50.799, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6462093862815884, | |
| "eval_loss": 0.6193148493766785, | |
| "eval_runtime": 0.5121, | |
| "eval_samples_per_second": 540.927, | |
| "eval_steps_per_second": 68.348, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.446111869031378e-05, | |
| "loss": 0.6759, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7075812274368231, | |
| "eval_loss": 0.6046352386474609, | |
| "eval_runtime": 0.5274, | |
| "eval_samples_per_second": 525.205, | |
| "eval_steps_per_second": 66.362, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7581227436823105, | |
| "eval_loss": 0.6365415453910828, | |
| "eval_runtime": 0.681, | |
| "eval_samples_per_second": 406.732, | |
| "eval_steps_per_second": 51.392, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7545126353790613, | |
| "eval_loss": 0.897487223148346, | |
| "eval_runtime": 0.6584, | |
| "eval_samples_per_second": 420.704, | |
| "eval_steps_per_second": 53.157, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 7.639836289222374e-06, | |
| "loss": 0.3194, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7581227436823105, | |
| "eval_loss": 1.2031357288360596, | |
| "eval_runtime": 0.5817, | |
| "eval_samples_per_second": 476.173, | |
| "eval_steps_per_second": 60.166, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7581227436823105, | |
| "eval_loss": 1.2942094802856445, | |
| "eval_runtime": 0.6818, | |
| "eval_samples_per_second": 406.254, | |
| "eval_steps_per_second": 51.332, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 1248, | |
| "total_flos": 1310293055692800.0, | |
| "train_loss": 0.42912168074876833, | |
| "train_runtime": 173.8183, | |
| "train_samples_per_second": 143.253, | |
| "train_steps_per_second": 8.975 | |
| } | |
| ], | |
| "max_steps": 1560, | |
| "num_train_epochs": 10, | |
| "total_flos": 1310293055692800.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |