| { | |
| "best_metric": 0.6021606922149658, | |
| "best_model_checkpoint": "mushrooms_image_detection/checkpoint-10945", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 10945, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 5.143975734710693, | |
| "learning_rate": 1.9173932996787516e-07, | |
| "loss": 0.6876, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 5.595354080200195, | |
| "learning_rate": 1.825608077099587e-07, | |
| "loss": 0.705, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 8.542109489440918, | |
| "learning_rate": 1.733822854520422e-07, | |
| "loss": 0.6908, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 5.49858283996582, | |
| "learning_rate": 1.6420376319412576e-07, | |
| "loss": 0.6763, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 5.070239067077637, | |
| "learning_rate": 1.5502524093620926e-07, | |
| "loss": 0.6807, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 3.0705084800720215, | |
| "learning_rate": 1.458467186782928e-07, | |
| "loss": 0.6879, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.836571455001831, | |
| "learning_rate": 1.366681964203763e-07, | |
| "loss": 0.69, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 7.066412448883057, | |
| "learning_rate": 1.2748967416245983e-07, | |
| "loss": 0.6808, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 5.1906328201293945, | |
| "learning_rate": 1.1831115190454337e-07, | |
| "loss": 0.6842, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 6.5265703201293945, | |
| "learning_rate": 1.0913262964662688e-07, | |
| "loss": 0.6911, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 8.26229190826416, | |
| "learning_rate": 9.995410738871042e-08, | |
| "loss": 0.6743, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 5.895538330078125, | |
| "learning_rate": 9.077558513079394e-08, | |
| "loss": 0.6734, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 9.030574798583984, | |
| "learning_rate": 8.159706287287747e-08, | |
| "loss": 0.6718, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 7.243845462799072, | |
| "learning_rate": 7.241854061496099e-08, | |
| "loss": 0.6698, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.422345161437988, | |
| "learning_rate": 6.32400183570445e-08, | |
| "loss": 0.6627, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 3.86919903755188, | |
| "learning_rate": 5.406149609912804e-08, | |
| "loss": 0.667, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 8.272964477539062, | |
| "learning_rate": 4.488297384121156e-08, | |
| "loss": 0.6542, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 7.44323205947876, | |
| "learning_rate": 3.5704451583295086e-08, | |
| "loss": 0.6729, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 3.3304898738861084, | |
| "learning_rate": 2.6525929325378617e-08, | |
| "loss": 0.6822, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 7.638361930847168, | |
| "learning_rate": 1.7347407067462138e-08, | |
| "loss": 0.6749, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 4.384842395782471, | |
| "learning_rate": 8.168884809545663e-09, | |
| "loss": 0.6733, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8896693506938496, | |
| "eval_loss": 0.6021606922149658, | |
| "eval_runtime": 3122.8795, | |
| "eval_samples_per_second": 74.764, | |
| "eval_steps_per_second": 9.346, | |
| "step": 10945 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10945, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 2.716308216840831e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |