| { | |
| "best_metric": 0.17856432497501373, | |
| "best_model_checkpoint": "bioformer8L_caption_reference_RTX6000/checkpoint-1230", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 2460, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4065040650406504, | |
| "grad_norm": 8.241156578063965, | |
| "learning_rate": 2.75609756097561e-05, | |
| "loss": 0.2277, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "grad_norm": 2.205291986465454, | |
| "learning_rate": 2.5121951219512197e-05, | |
| "loss": 0.177, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9276147633166343, | |
| "eval_f1": 0.9222332155896071, | |
| "eval_f1_medical": 0.9426906265177271, | |
| "eval_f1_non_medical": 0.9017758046614872, | |
| "eval_loss": 0.17856432497501373, | |
| "eval_precision": 0.9277090451077986, | |
| "eval_recall": 0.9177374028102576, | |
| "eval_runtime": 37.413, | |
| "eval_samples_per_second": 261.433, | |
| "eval_steps_per_second": 8.179, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 3.5154032707214355, | |
| "learning_rate": 2.2682926829268295e-05, | |
| "loss": 0.1466, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "grad_norm": 3.8458199501037598, | |
| "learning_rate": 2.024390243902439e-05, | |
| "loss": 0.1232, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.929352826909314, | |
| "eval_f1": 0.9248187483559904, | |
| "eval_f1_medical": 0.9432816219322006, | |
| "eval_f1_non_medical": 0.9063558747797804, | |
| "eval_loss": 0.20747286081314087, | |
| "eval_precision": 0.9255804539078865, | |
| "eval_recall": 0.9240820913256004, | |
| "eval_runtime": 37.4622, | |
| "eval_samples_per_second": 261.09, | |
| "eval_steps_per_second": 8.168, | |
| "step": 2460 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6162603809728128.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |