| { | |
| "best_metric": 0.6494115591049194, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qnli_128/checkpoint-819", | |
| "epoch": 6.0, | |
| "global_step": 4914, | |
| "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": 0.6774, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6135822807980963, | |
| "eval_loss": 0.6494115591049194, | |
| "eval_runtime": 12.0846, | |
| "eval_samples_per_second": 452.061, | |
| "eval_steps_per_second": 3.558, | |
| "step": 819 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6378, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6055280981145891, | |
| "eval_loss": 0.650793194770813, | |
| "eval_runtime": 12.0869, | |
| "eval_samples_per_second": 451.976, | |
| "eval_steps_per_second": 3.558, | |
| "step": 1638 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6148, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6062602965403624, | |
| "eval_loss": 0.6577999591827393, | |
| "eval_runtime": 12.096, | |
| "eval_samples_per_second": 451.636, | |
| "eval_steps_per_second": 3.555, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5979, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6060772469339191, | |
| "eval_loss": 0.6590101718902588, | |
| "eval_runtime": 12.1006, | |
| "eval_samples_per_second": 451.464, | |
| "eval_steps_per_second": 3.554, | |
| "step": 3276 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5851, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5927146256635548, | |
| "eval_loss": 0.6761044859886169, | |
| "eval_runtime": 12.0743, | |
| "eval_samples_per_second": 452.449, | |
| "eval_steps_per_second": 3.561, | |
| "step": 4095 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5743, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5978400146439685, | |
| "eval_loss": 0.6981686949729919, | |
| "eval_runtime": 12.0788, | |
| "eval_samples_per_second": 452.279, | |
| "eval_steps_per_second": 3.56, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 4914, | |
| "total_flos": 1.4979041505312768e+16, | |
| "train_loss": 0.614554194449214, | |
| "train_runtime": 4192.3502, | |
| "train_samples_per_second": 1249.216, | |
| "train_steps_per_second": 9.768 | |
| } | |
| ], | |
| "max_steps": 40950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.4979041505312768e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |