| { | |
| "best_metric": 0.022927358746528625, | |
| "best_model_checkpoint": "deepfake_vs_real_image_detection/checkpoint-3571", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 3571, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.26625385880470276, | |
| "learning_rate": 8.721953990343652e-07, | |
| "loss": 0.0944, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.4247075319290161, | |
| "learning_rate": 7.301902868503266e-07, | |
| "loss": 0.0715, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 7.891324520111084, | |
| "learning_rate": 5.881851746662879e-07, | |
| "loss": 0.069, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 6.847803115844727, | |
| "learning_rate": 4.461800624822493e-07, | |
| "loss": 0.0658, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.0988214015960693, | |
| "learning_rate": 3.041749502982107e-07, | |
| "loss": 0.0632, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 6.4862895011901855, | |
| "learning_rate": 1.621698381141721e-07, | |
| "loss": 0.0698, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.6512278914451599, | |
| "learning_rate": 2.0164725930133484e-08, | |
| "loss": 0.0587, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9926865456073318, | |
| "eval_loss": 0.022927358746528625, | |
| "eval_runtime": 773.2123, | |
| "eval_samples_per_second": 98.499, | |
| "eval_steps_per_second": 12.314, | |
| "step": 3571 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3571, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 8.852762385560605e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |