| { | |
| "best_metric": 0.6895958781242371, | |
| "best_model_checkpoint": "hBERTv1_rte/checkpoint-10", | |
| "epoch": 6.0, | |
| "global_step": 60, | |
| "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": 0.7247, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5451263537906137, | |
| "eval_loss": 0.6895958781242371, | |
| "eval_runtime": 0.3617, | |
| "eval_samples_per_second": 765.769, | |
| "eval_steps_per_second": 5.529, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7046, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7013799548149109, | |
| "eval_runtime": 0.363, | |
| "eval_samples_per_second": 763.011, | |
| "eval_steps_per_second": 5.509, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6934, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6982845067977905, | |
| "eval_runtime": 0.3662, | |
| "eval_samples_per_second": 756.401, | |
| "eval_steps_per_second": 5.461, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6846, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5126353790613718, | |
| "eval_loss": 0.7091580033302307, | |
| "eval_runtime": 0.3627, | |
| "eval_samples_per_second": 763.647, | |
| "eval_steps_per_second": 5.514, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6853, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5126353790613718, | |
| "eval_loss": 0.713951826095581, | |
| "eval_runtime": 0.3618, | |
| "eval_samples_per_second": 765.669, | |
| "eval_steps_per_second": 5.528, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6152, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.49097472924187724, | |
| "eval_loss": 0.8229976892471313, | |
| "eval_runtime": 0.3616, | |
| "eval_samples_per_second": 766.085, | |
| "eval_steps_per_second": 5.531, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 60, | |
| "total_flos": 1887086894383104.0, | |
| "train_loss": 0.6846395254135131, | |
| "train_runtime": 109.2886, | |
| "train_samples_per_second": 1139.185, | |
| "train_steps_per_second": 4.575 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 50, | |
| "total_flos": 1887086894383104.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |