| { | |
| "best_metric": 0.47375271850377065, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-4\\checkpoint-4276", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4276, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 6.042296562538321e-06, | |
| "loss": 0.606, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 5.242204327032718e-06, | |
| "loss": 0.5464, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.442112091527114e-06, | |
| "loss": 0.5152, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.642019856021509e-06, | |
| "loss": 0.5153, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5116811394691467, | |
| "eval_matthews_correlation": 0.46741180953359257, | |
| "eval_runtime": 5.9325, | |
| "eval_samples_per_second": 175.81, | |
| "eval_steps_per_second": 11.125, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 2.8419276205159054e-06, | |
| "loss": 0.4465, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.041835385010301e-06, | |
| "loss": 0.4589, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.2417431495046974e-06, | |
| "loss": 0.527, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.4165091399909343e-07, | |
| "loss": 0.5133, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7559528946876526, | |
| "eval_matthews_correlation": 0.47375271850377065, | |
| "eval_runtime": 5.972, | |
| "eval_samples_per_second": 174.648, | |
| "eval_steps_per_second": 11.052, | |
| "step": 4276 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4276, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 65223117962472.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 6.842388798043926e-06, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 11 | |
| } | |
| } | |