| { | |
| "best_metric": 0.6930862069129944, | |
| "best_model_checkpoint": "hBERTv1_data_aug_qnli/checkpoint-16604", | |
| "epoch": 6.0, | |
| "global_step": 99624, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9000421585160206e-05, | |
| "loss": 0.6933, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930862069129944, | |
| "eval_runtime": 6.9764, | |
| "eval_samples_per_second": 783.068, | |
| "eval_steps_per_second": 3.153, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800090339677186e-05, | |
| "loss": 0.6932, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931231617927551, | |
| "eval_runtime": 7.005, | |
| "eval_samples_per_second": 779.871, | |
| "eval_steps_per_second": 3.141, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700138520838353e-05, | |
| "loss": 0.6932, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930967569351196, | |
| "eval_runtime": 6.9994, | |
| "eval_samples_per_second": 780.494, | |
| "eval_steps_per_second": 3.143, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600186701999518e-05, | |
| "loss": 0.6932, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931020617485046, | |
| "eval_runtime": 6.9797, | |
| "eval_samples_per_second": 782.698, | |
| "eval_steps_per_second": 3.152, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.500234883160684e-05, | |
| "loss": 0.6932, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931495070457458, | |
| "eval_runtime": 7.0501, | |
| "eval_samples_per_second": 774.887, | |
| "eval_steps_per_second": 3.121, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4002830643218504e-05, | |
| "loss": 0.6932, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930967569351196, | |
| "eval_runtime": 6.9694, | |
| "eval_samples_per_second": 783.852, | |
| "eval_steps_per_second": 3.157, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 99624, | |
| "total_flos": 3.221318655610454e+18, | |
| "train_loss": 0.693251175200002, | |
| "train_runtime": 91175.0516, | |
| "train_samples_per_second": 2330.962, | |
| "train_steps_per_second": 9.106 | |
| } | |
| ], | |
| "max_steps": 830200, | |
| "num_train_epochs": 50, | |
| "total_flos": 3.221318655610454e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |