| { | |
| "best_metric": 2.5123291015625, | |
| "best_model_checkpoint": "cat_breed_image_detection/checkpoint-3811", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 3811, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.980210304260254, | |
| "learning_rate": 2.641052911459718e-07, | |
| "loss": 2.5248, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.637603282928467, | |
| "learning_rate": 2.2422228130816273e-07, | |
| "loss": 2.4992, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.8118417263031006, | |
| "learning_rate": 1.8433927147035363e-07, | |
| "loss": 2.4915, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 3.189683437347412, | |
| "learning_rate": 1.444562616325445e-07, | |
| "loss": 2.4989, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 2.384906768798828, | |
| "learning_rate": 1.0457325179473543e-07, | |
| "loss": 2.4949, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 3.004359006881714, | |
| "learning_rate": 6.469024195692635e-08, | |
| "loss": 2.4943, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 2.5521860122680664, | |
| "learning_rate": 2.4807232119117254e-08, | |
| "loss": 2.4874, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.41788455860643636, | |
| "eval_loss": 2.5123291015625, | |
| "eval_runtime": 307.8834, | |
| "eval_samples_per_second": 88.007, | |
| "eval_steps_per_second": 2.751, | |
| "step": 3811 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3811, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 1.8905297684573454e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |