| { | |
| "best_metric": 0.5410182476043701, | |
| "best_model_checkpoint": "AI-vs-Deepfake-vs-Real/checkpoint-752", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 752, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8195, | |
| "eval_loss": 0.9633184671401978, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 92.615, | |
| "eval_samples_per_second": 43.19, | |
| "eval_steps_per_second": 5.399, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.96125, | |
| "eval_loss": 0.7438851594924927, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 93.1265, | |
| "eval_samples_per_second": 42.952, | |
| "eval_steps_per_second": 5.369, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 1.5466415882110596, | |
| "learning_rate": 7.179487179487179e-07, | |
| "loss": 0.8098, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.974, | |
| "eval_loss": 0.5879941582679749, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 92.9279, | |
| "eval_samples_per_second": 43.044, | |
| "eval_steps_per_second": 5.381, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.975, | |
| "eval_loss": 0.5410182476043701, | |
| "eval_model_preparation_time": 0.0043, | |
| "eval_runtime": 93.7304, | |
| "eval_samples_per_second": 42.676, | |
| "eval_steps_per_second": 5.334, | |
| "step": 752 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 752, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.8595144488822129e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |