| { | |
| "best_metric": 0.6926400661468506, | |
| "best_model_checkpoint": "mobilebert_add_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": 0.6934, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6926400661468506, | |
| "eval_runtime": 0.591, | |
| "eval_samples_per_second": 468.723, | |
| "eval_steps_per_second": 5.076, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6933, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6929883360862732, | |
| "eval_runtime": 0.5159, | |
| "eval_samples_per_second": 536.905, | |
| "eval_steps_per_second": 5.815, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6931, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4657039711191336, | |
| "eval_loss": 0.6933187246322632, | |
| "eval_runtime": 0.5114, | |
| "eval_samples_per_second": 541.605, | |
| "eval_steps_per_second": 5.866, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6932, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6928666830062866, | |
| "eval_runtime": 0.5061, | |
| "eval_samples_per_second": 547.347, | |
| "eval_steps_per_second": 5.928, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6931, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6942214965820312, | |
| "eval_runtime": 0.4994, | |
| "eval_samples_per_second": 554.668, | |
| "eval_steps_per_second": 6.007, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6926, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6929604411125183, | |
| "eval_runtime": 0.5098, | |
| "eval_samples_per_second": 543.359, | |
| "eval_steps_per_second": 5.885, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 120, | |
| "total_flos": 356657204822016.0, | |
| "train_loss": 0.6931220054626465, | |
| "train_runtime": 130.999, | |
| "train_samples_per_second": 950.389, | |
| "train_steps_per_second": 7.634 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 50, | |
| "total_flos": 356657204822016.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |