| { | |
| "best_metric": 0.8676160051978992, | |
| "best_model_checkpoint": "/kaggle/working/sweep_14/checkpoint-174", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 174, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 11.585850715637207, | |
| "learning_rate": 2.2822085889570554e-05, | |
| "loss": 0.2943, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.754601226993865, | |
| "eval_f1_macro": 0.711504424778761, | |
| "eval_loss": 0.1252642422914505, | |
| "eval_precision_macro": 0.7120721554116559, | |
| "eval_recall_macro": 0.8361436950146628, | |
| "eval_runtime": 0.9487, | |
| "eval_samples_per_second": 171.813, | |
| "eval_steps_per_second": 6.324, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 6.7062201499938965, | |
| "learning_rate": 1.3619631901840491e-05, | |
| "loss": 0.1982, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.852760736196319, | |
| "eval_f1_macro": 0.807556080283353, | |
| "eval_loss": 0.08962057530879974, | |
| "eval_precision_macro": 0.7784734133790738, | |
| "eval_recall_macro": 0.8967497556207233, | |
| "eval_runtime": 0.9489, | |
| "eval_samples_per_second": 171.781, | |
| "eval_steps_per_second": 6.323, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.586206896551724, | |
| "grad_norm": 6.61816930770874, | |
| "learning_rate": 4.41717791411043e-06, | |
| "loss": 0.1052, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9079754601226994, | |
| "eval_f1_macro": 0.8676160051978992, | |
| "eval_loss": 0.07852204144001007, | |
| "eval_precision_macro": 0.8369736324281779, | |
| "eval_recall_macro": 0.9184995112414467, | |
| "eval_runtime": 0.9267, | |
| "eval_samples_per_second": 175.897, | |
| "eval_steps_per_second": 6.475, | |
| "step": 174 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 174, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 2, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 482124676077168.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |