| { | |
| "best_metric": 0.543909089281463, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-2138", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 5345, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.0681209971615274e-05, | |
| "loss": 0.5488, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.8546926176814934e-05, | |
| "loss": 0.498, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.48384082317352295, | |
| "eval_matthews_correlation": 0.4611631351812362, | |
| "eval_runtime": 0.7492, | |
| "eval_samples_per_second": 1392.127, | |
| "eval_steps_per_second": 88.092, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.6412642382014598e-05, | |
| "loss": 0.3532, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.4278358587214258e-05, | |
| "loss": 0.354, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5868804454803467, | |
| "eval_matthews_correlation": 0.543909089281463, | |
| "eval_runtime": 0.7926, | |
| "eval_samples_per_second": 1315.987, | |
| "eval_steps_per_second": 83.274, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.214407479241392e-05, | |
| "loss": 0.2699, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.0009790997613584e-05, | |
| "loss": 0.2462, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7930343747138977, | |
| "eval_matthews_correlation": 0.5340667882909217, | |
| "eval_runtime": 0.7474, | |
| "eval_samples_per_second": 1395.496, | |
| "eval_steps_per_second": 88.306, | |
| "step": 3207 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 7.875507202813246e-06, | |
| "loss": 0.1931, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 5.741223408012909e-06, | |
| "loss": 0.1671, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.0090364217758179, | |
| "eval_matthews_correlation": 0.5279532549901665, | |
| "eval_runtime": 1.1191, | |
| "eval_samples_per_second": 931.964, | |
| "eval_steps_per_second": 58.974, | |
| "step": 4276 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 3.6069396132125703e-06, | |
| "loss": 0.1374, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 1.472655818412233e-06, | |
| "loss": 0.097, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.1014072895050049, | |
| "eval_matthews_correlation": 0.5211120728046958, | |
| "eval_runtime": 0.7198, | |
| "eval_samples_per_second": 1448.925, | |
| "eval_steps_per_second": 91.687, | |
| "step": 5345 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5345, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 188320776692052.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 2.281549376641561e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 24 | |
| } | |
| } | |