| { | |
| "best_metric": 0.6928680539131165, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_rte_256/checkpoint-20", | |
| "epoch": 6.0, | |
| "global_step": 120, | |
| "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": 0.6937, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6928680539131165, | |
| "eval_runtime": 0.5064, | |
| "eval_samples_per_second": 547.033, | |
| "eval_steps_per_second": 5.925, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6938, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6928784847259521, | |
| "eval_runtime": 0.5193, | |
| "eval_samples_per_second": 533.424, | |
| "eval_steps_per_second": 5.777, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6931, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5126353790613718, | |
| "eval_loss": 0.6931412220001221, | |
| "eval_runtime": 0.5062, | |
| "eval_samples_per_second": 547.26, | |
| "eval_steps_per_second": 5.927, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6932, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4693140794223827, | |
| "eval_loss": 0.693750262260437, | |
| "eval_runtime": 0.5127, | |
| "eval_samples_per_second": 540.292, | |
| "eval_steps_per_second": 5.852, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.693, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6949868202209473, | |
| "eval_runtime": 0.5398, | |
| "eval_samples_per_second": 513.2, | |
| "eval_steps_per_second": 5.558, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6921, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.51985559566787, | |
| "eval_loss": 0.6932752728462219, | |
| "eval_runtime": 0.519, | |
| "eval_samples_per_second": 533.7, | |
| "eval_steps_per_second": 5.78, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 120, | |
| "total_flos": 385081701040128.0, | |
| "train_loss": 0.6931396325429281, | |
| "train_runtime": 122.3877, | |
| "train_samples_per_second": 1017.259, | |
| "train_steps_per_second": 8.171 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 50, | |
| "total_flos": 385081701040128.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |