| { | |
| "best_metric": 0.6179953217506409, | |
| "best_model_checkpoint": "hBERTv1_cola/checkpoint-34", | |
| "epoch": 6.0, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6212, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6179953217506409, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.3043, | |
| "eval_samples_per_second": 799.657, | |
| "eval_steps_per_second": 3.833, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6118, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6210665702819824, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.3002, | |
| "eval_samples_per_second": 802.193, | |
| "eval_steps_per_second": 3.846, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6088, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6251837611198425, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.3073, | |
| "eval_samples_per_second": 797.82, | |
| "eval_steps_per_second": 3.825, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6103, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6181675791740417, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.307, | |
| "eval_samples_per_second": 798.007, | |
| "eval_steps_per_second": 3.826, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6107, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6210399270057678, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.3053, | |
| "eval_samples_per_second": 799.078, | |
| "eval_steps_per_second": 3.831, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6081, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6212285757064819, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 1.3083, | |
| "eval_samples_per_second": 797.202, | |
| "eval_steps_per_second": 3.822, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 204, | |
| "total_flos": 6480513749483520.0, | |
| "train_loss": 0.5082857468548942, | |
| "train_runtime": 207.6378, | |
| "train_samples_per_second": 2059.114, | |
| "train_steps_per_second": 8.187 | |
| } | |
| ], | |
| "max_steps": 1700, | |
| "num_train_epochs": 50, | |
| "total_flos": 6480513749483520.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |