| { |
| "best_metric": 2.328033447265625, |
| "best_model_checkpoint": "hBERTv1_data_aug_rte/checkpoint-568", |
| "epoch": 6.0, |
| "global_step": 3408, |
| "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": 0.2316, |
| "step": 568 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.51985559566787, |
| "eval_loss": 2.328033447265625, |
| "eval_runtime": 0.3648, |
| "eval_samples_per_second": 759.418, |
| "eval_steps_per_second": 5.483, |
| "step": 568 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 4.8e-05, |
| "loss": 0.0234, |
| "step": 1136 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5270758122743683, |
| "eval_loss": 3.0592288970947266, |
| "eval_runtime": 0.365, |
| "eval_samples_per_second": 758.809, |
| "eval_steps_per_second": 5.479, |
| "step": 1136 |
| }, |
| { |
| "epoch": 3.0, |
| "learning_rate": 4.7e-05, |
| "loss": 0.0106, |
| "step": 1704 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.5054151624548736, |
| "eval_loss": 3.697169542312622, |
| "eval_runtime": 0.3629, |
| "eval_samples_per_second": 763.365, |
| "eval_steps_per_second": 5.512, |
| "step": 1704 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.0067, |
| "step": 2272 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.47653429602888087, |
| "eval_loss": 3.2471261024475098, |
| "eval_runtime": 0.3643, |
| "eval_samples_per_second": 760.433, |
| "eval_steps_per_second": 5.49, |
| "step": 2272 |
| }, |
| { |
| "epoch": 5.0, |
| "learning_rate": 4.5e-05, |
| "loss": 0.0051, |
| "step": 2840 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.5090252707581228, |
| "eval_loss": 3.3428096771240234, |
| "eval_runtime": 0.3658, |
| "eval_samples_per_second": 757.254, |
| "eval_steps_per_second": 5.468, |
| "step": 2840 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.0037, |
| "step": 3408 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.5054151624548736, |
| "eval_loss": 3.9613287448883057, |
| "eval_runtime": 0.3657, |
| "eval_samples_per_second": 757.365, |
| "eval_steps_per_second": 5.468, |
| "step": 3408 |
| }, |
| { |
| "epoch": 6.0, |
| "step": 3408, |
| "total_flos": 1.1019223499695718e+17, |
| "train_loss": 0.04685182029932317, |
| "train_runtime": 3272.6204, |
| "train_samples_per_second": 2221.431, |
| "train_steps_per_second": 8.678 |
| } |
| ], |
| "max_steps": 28400, |
| "num_train_epochs": 50, |
| "total_flos": 1.1019223499695718e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
