| { |
| "best_metric": 0.7326732673267325, |
| "best_model_checkpoint": "final_models/structroberta_sx2_final/finetune/qnli/checkpoint-2000", |
| "epoch": 5.46448087431694, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.09, |
| "eval_accuracy": 0.6736658215522766, |
| "eval_f1": 0.681468830059778, |
| "eval_loss": 0.6080768704414368, |
| "eval_mcc": 0.34894047189705113, |
| "eval_runtime": 4.8444, |
| "eval_samples_per_second": 471.881, |
| "eval_steps_per_second": 59.037, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.37, |
| "learning_rate": 4.316939890710383e-05, |
| "loss": 0.6434, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.19, |
| "eval_accuracy": 0.7060367465019226, |
| "eval_f1": 0.6759884281581485, |
| "eval_loss": 0.5899884700775146, |
| "eval_mcc": 0.4424515611165257, |
| "eval_runtime": 4.7984, |
| "eval_samples_per_second": 476.406, |
| "eval_steps_per_second": 59.603, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.73, |
| "learning_rate": 3.633879781420765e-05, |
| "loss": 0.5162, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.28, |
| "eval_accuracy": 0.7314085960388184, |
| "eval_f1": 0.718348623853211, |
| "eval_loss": 0.5835453271865845, |
| "eval_mcc": 0.47941652405255153, |
| "eval_runtime": 4.8413, |
| "eval_samples_per_second": 472.185, |
| "eval_steps_per_second": 59.075, |
| "step": 1200 |
| }, |
| { |
| "epoch": 4.1, |
| "learning_rate": 2.9508196721311478e-05, |
| "loss": 0.3707, |
| "step": 1500 |
| }, |
| { |
| "epoch": 4.37, |
| "eval_accuracy": 0.7397200465202332, |
| "eval_f1": 0.7318611987381702, |
| "eval_loss": 0.6178868412971497, |
| "eval_mcc": 0.4917193731545861, |
| "eval_runtime": 4.802, |
| "eval_samples_per_second": 476.055, |
| "eval_steps_per_second": 59.559, |
| "step": 1600 |
| }, |
| { |
| "epoch": 5.46, |
| "learning_rate": 2.2677595628415303e-05, |
| "loss": 0.255, |
| "step": 2000 |
| }, |
| { |
| "epoch": 5.46, |
| "eval_accuracy": 0.7401574850082397, |
| "eval_f1": 0.7326732673267325, |
| "eval_loss": 0.7336315512657166, |
| "eval_mcc": 0.4922779179551397, |
| "eval_runtime": 4.8165, |
| "eval_samples_per_second": 474.623, |
| "eval_steps_per_second": 59.38, |
| "step": 2000 |
| } |
| ], |
| "max_steps": 3660, |
| "num_train_epochs": 10, |
| "total_flos": 2.198357346845952e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
