| { | |
| "best_metric": 0.021508002653717995, | |
| "best_model_checkpoint": "deepfake_vs_real_image_detection/checkpoint-5653", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 5653, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.0086122751235962, | |
| "learning_rate": 4.598429412814564e-06, | |
| "loss": 0.0287, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.11366910487413406, | |
| "learning_rate": 4.152239871497413e-06, | |
| "loss": 0.0292, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.04216881841421127, | |
| "learning_rate": 3.706050330180261e-06, | |
| "loss": 0.0273, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.0077363960444927216, | |
| "learning_rate": 3.2598607888631096e-06, | |
| "loss": 0.0288, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.012885607779026031, | |
| "learning_rate": 2.8136712475459575e-06, | |
| "loss": 0.0285, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.006761148571968079, | |
| "learning_rate": 2.367481706228806e-06, | |
| "loss": 0.0246, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.020215608179569244, | |
| "learning_rate": 1.9212921649116544e-06, | |
| "loss": 0.026, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.03672046586871147, | |
| "learning_rate": 1.4751026235945032e-06, | |
| "loss": 0.0254, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.4788285493850708, | |
| "learning_rate": 1.0289130822773514e-06, | |
| "loss": 0.0262, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.04541263356804848, | |
| "learning_rate": 5.827235409601999e-07, | |
| "loss": 0.028, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.007974252104759216, | |
| "learning_rate": 1.3653399964304838e-07, | |
| "loss": 0.0253, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9934880789833, | |
| "eval_loss": 0.021508002653717995, | |
| "eval_runtime": 96.1411, | |
| "eval_samples_per_second": 99.031, | |
| "eval_steps_per_second": 12.388, | |
| "step": 5653 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5653, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 1.4016828573477016e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |