| { | |
| "best_metric": NaN, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_pretrain_sst2/checkpoint-527", | |
| "epoch": 6.0, | |
| "global_step": 3162, | |
| "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": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5233, | |
| "eval_samples_per_second": 572.445, | |
| "eval_steps_per_second": 4.595, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5073, | |
| "eval_samples_per_second": 578.515, | |
| "eval_steps_per_second": 4.644, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.0, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5082, | |
| "eval_samples_per_second": 578.154, | |
| "eval_steps_per_second": 4.641, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5023, | |
| "eval_samples_per_second": 580.451, | |
| "eval_steps_per_second": 4.66, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5058, | |
| "eval_samples_per_second": 579.081, | |
| "eval_steps_per_second": 4.649, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.5043, | |
| "eval_samples_per_second": 579.69, | |
| "eval_steps_per_second": 4.653, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 3162, | |
| "total_flos": 1.2116275614449664e+16, | |
| "train_loss": 0.0, | |
| "train_runtime": 1714.5809, | |
| "train_samples_per_second": 1964.008, | |
| "train_steps_per_second": 15.368 | |
| } | |
| ], | |
| "max_steps": 26350, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.2116275614449664e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |