| { | |
| "best_metric": 0.6926150321960449, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_rte_128/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": 0.6935, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6926150321960449, | |
| "eval_runtime": 0.6362, | |
| "eval_samples_per_second": 435.421, | |
| "eval_steps_per_second": 4.716, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6934, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6930063366889954, | |
| "eval_runtime": 0.6364, | |
| "eval_samples_per_second": 435.283, | |
| "eval_steps_per_second": 4.714, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6931, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4981949458483754, | |
| "eval_loss": 0.6932486891746521, | |
| "eval_runtime": 0.6297, | |
| "eval_samples_per_second": 439.893, | |
| "eval_steps_per_second": 4.764, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6932, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5342960288808665, | |
| "eval_loss": 0.6928855180740356, | |
| "eval_runtime": 0.635, | |
| "eval_samples_per_second": 436.245, | |
| "eval_steps_per_second": 4.725, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6929, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6944915056228638, | |
| "eval_runtime": 0.6203, | |
| "eval_samples_per_second": 446.572, | |
| "eval_steps_per_second": 4.837, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6921, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.51985559566787, | |
| "eval_loss": 0.6929416656494141, | |
| "eval_runtime": 0.6349, | |
| "eval_samples_per_second": 436.304, | |
| "eval_steps_per_second": 4.725, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 120, | |
| "total_flos": 356088834686976.0, | |
| "train_loss": 0.6930288791656494, | |
| "train_runtime": 153.3083, | |
| "train_samples_per_second": 812.089, | |
| "train_steps_per_second": 6.523 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 50, | |
| "total_flos": 356088834686976.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |