{ "best_metric": 0.5177230834960938, "best_model_checkpoint": "vit-base-patch16-224-in21k-rotated-dungeons-v3/checkpoint-100", "epoch": 15.0, "eval_steps": 100, "global_step": 120, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.25, "grad_norm": 2.4491872787475586, "learning_rate": 0.00018333333333333334, "loss": 1.7253, "step": 10 }, { "epoch": 2.5, "grad_norm": 2.151639461517334, "learning_rate": 0.0001666666666666667, "loss": 1.2096, "step": 20 }, { "epoch": 3.75, "grad_norm": 1.7387666702270508, "learning_rate": 0.00015000000000000001, "loss": 0.8385, "step": 30 }, { "epoch": 5.0, "grad_norm": 1.5225492715835571, "learning_rate": 0.00013333333333333334, "loss": 0.4285, "step": 40 }, { "epoch": 6.25, "grad_norm": 0.6194241046905518, "learning_rate": 0.00011666666666666668, "loss": 0.322, "step": 50 }, { "epoch": 7.5, "grad_norm": 0.9321300387382507, "learning_rate": 0.0001, "loss": 0.1986, "step": 60 }, { "epoch": 8.75, "grad_norm": 0.4772986173629761, "learning_rate": 8.333333333333334e-05, "loss": 0.1683, "step": 70 }, { "epoch": 10.0, "grad_norm": 0.4312579929828644, "learning_rate": 6.666666666666667e-05, "loss": 0.1219, "step": 80 }, { "epoch": 11.25, "grad_norm": 0.34619101881980896, "learning_rate": 5e-05, "loss": 0.1029, "step": 90 }, { "epoch": 12.5, "grad_norm": 0.330285906791687, "learning_rate": 3.3333333333333335e-05, "loss": 0.1009, "step": 100 }, { "epoch": 12.5, "eval_accuracy": 0.875, "eval_loss": 0.5177230834960938, "eval_runtime": 0.0932, "eval_samples_per_second": 85.814, "eval_steps_per_second": 10.727, "step": 100 }, { "epoch": 13.75, "grad_norm": 0.5117152333259583, "learning_rate": 1.6666666666666667e-05, "loss": 0.0982, "step": 110 }, { "epoch": 15.0, "grad_norm": 0.43090176582336426, "learning_rate": 0.0, "loss": 0.0872, "step": 120 }, { "epoch": 15.0, "step": 120, "total_flos": 4.64972773773312e+16, "train_loss": 0.45015646318594615, "train_runtime": 30.2968, "train_samples_per_second": 19.804, "train_steps_per_second": 3.961 } ], "logging_steps": 10, "max_steps": 120, "num_input_tokens_seen": 0, "num_train_epochs": 15, "save_steps": 100, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 4.64972773773312e+16, "train_batch_size": 5, "trial_name": null, "trial_params": null }