| { | |
| "best_metric": NaN, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_pretrain_stsb/checkpoint-45", | |
| "epoch": 6.0, | |
| "global_step": 270, | |
| "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": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 3.2792, | |
| "eval_samples_per_second": 457.422, | |
| "eval_steps_per_second": 3.659, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 4.2807, | |
| "eval_samples_per_second": 350.411, | |
| "eval_steps_per_second": 2.803, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.0, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 3.2608, | |
| "eval_samples_per_second": 460.015, | |
| "eval_steps_per_second": 3.68, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 4.301, | |
| "eval_samples_per_second": 348.759, | |
| "eval_steps_per_second": 2.79, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 3.2605, | |
| "eval_samples_per_second": 460.047, | |
| "eval_steps_per_second": 3.68, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": NaN, | |
| "eval_mse": NaN, | |
| "eval_runtime": 4.2804, | |
| "eval_samples_per_second": 350.434, | |
| "eval_steps_per_second": 2.803, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 270, | |
| "total_flos": 1034234131120128.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 274.6362, | |
| "train_samples_per_second": 1046.657, | |
| "train_steps_per_second": 8.193 | |
| } | |
| ], | |
| "max_steps": 2250, | |
| "num_train_epochs": 50, | |
| "total_flos": 1034234131120128.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |