| { | |
| "best_metric": 0.6172745823860168, | |
| "best_model_checkpoint": "hBERTv2_new_pretrain_cola/checkpoint-201", | |
| "epoch": 8.0, | |
| "global_step": 536, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.6294, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6236423850059509, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7907, | |
| "eval_samples_per_second": 582.451, | |
| "eval_steps_per_second": 5.026, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.6169, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6311982870101929, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7917, | |
| "eval_samples_per_second": 582.113, | |
| "eval_steps_per_second": 5.023, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.6115, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6172745823860168, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7888, | |
| "eval_samples_per_second": 583.068, | |
| "eval_steps_per_second": 5.031, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.6372, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.620063841342926, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7867, | |
| "eval_samples_per_second": 583.759, | |
| "eval_steps_per_second": 5.037, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6087, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6216873526573181, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7924, | |
| "eval_samples_per_second": 581.886, | |
| "eval_steps_per_second": 5.021, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.6086, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.624801516532898, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7896, | |
| "eval_samples_per_second": 582.822, | |
| "eval_steps_per_second": 5.029, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.6113, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.6282826662063599, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.7883, | |
| "eval_samples_per_second": 583.234, | |
| "eval_steps_per_second": 5.033, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 0.6109, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.620032012462616, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.8037, | |
| "eval_samples_per_second": 578.242, | |
| "eval_steps_per_second": 4.99, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 536, | |
| "total_flos": 1.0010769903058944e+16, | |
| "train_loss": 0.6167977461174353, | |
| "train_runtime": 414.5842, | |
| "train_samples_per_second": 1031.274, | |
| "train_steps_per_second": 8.08 | |
| } | |
| ], | |
| "max_steps": 3350, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.0010769903058944e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |