| { | |
| "best_metric": 1.0619962215423584, | |
| "best_model_checkpoint": "cat_breed_image_detection/checkpoint-7623", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 7623, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19677292404565133, | |
| "grad_norm": 12.01923656463623, | |
| "learning_rate": 7.524626964214973e-07, | |
| "loss": 1.0123, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39354584809130266, | |
| "grad_norm": 11.533829689025879, | |
| "learning_rate": 6.996434702231612e-07, | |
| "loss": 1.0073, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5903187721369539, | |
| "grad_norm": 12.163763999938965, | |
| "learning_rate": 6.46824244024825e-07, | |
| "loss": 1.004, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7870916961826053, | |
| "grad_norm": 8.848891258239746, | |
| "learning_rate": 5.940050178264888e-07, | |
| "loss": 1.002, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9838646202282566, | |
| "grad_norm": 8.833011627197266, | |
| "learning_rate": 5.411857916281527e-07, | |
| "loss": 0.9938, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7094405078240331, | |
| "eval_loss": 1.0737584829330444, | |
| "eval_model_preparation_time": 0.0066, | |
| "eval_runtime": 1208.9637, | |
| "eval_samples_per_second": 89.65, | |
| "eval_steps_per_second": 2.802, | |
| "step": 2541 | |
| }, | |
| { | |
| "epoch": 1.1806375442739079, | |
| "grad_norm": 8.953508377075195, | |
| "learning_rate": 4.883665654298165e-07, | |
| "loss": 0.989, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3774104683195592, | |
| "grad_norm": 10.335805892944336, | |
| "learning_rate": 4.355473392314803e-07, | |
| "loss": 0.9866, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5741833923652107, | |
| "grad_norm": 11.203774452209473, | |
| "learning_rate": 3.8272811303314405e-07, | |
| "loss": 0.9972, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.770956316410862, | |
| "grad_norm": 8.466325759887695, | |
| "learning_rate": 3.2990888683480783e-07, | |
| "loss": 0.9824, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9677292404565132, | |
| "grad_norm": 7.2133331298828125, | |
| "learning_rate": 2.7708966063647166e-07, | |
| "loss": 0.9799, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7116087245349867, | |
| "eval_loss": 1.0655748844146729, | |
| "eval_model_preparation_time": 0.0066, | |
| "eval_runtime": 1248.8826, | |
| "eval_samples_per_second": 86.785, | |
| "eval_steps_per_second": 2.712, | |
| "step": 5082 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 6.7830491065979, | |
| "learning_rate": 2.2427043443813546e-07, | |
| "loss": 0.9948, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3612750885478158, | |
| "grad_norm": 10.073229789733887, | |
| "learning_rate": 1.714512082397993e-07, | |
| "loss": 0.966, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.558048012593467, | |
| "grad_norm": 8.630289077758789, | |
| "learning_rate": 1.186319820414631e-07, | |
| "loss": 0.9904, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7548209366391183, | |
| "grad_norm": 9.734028816223145, | |
| "learning_rate": 6.58127558431269e-08, | |
| "loss": 0.9671, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9515938606847696, | |
| "grad_norm": 13.89803695678711, | |
| "learning_rate": 1.2993529644790705e-08, | |
| "loss": 0.9834, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.712356067316209, | |
| "eval_loss": 1.0619962215423584, | |
| "eval_model_preparation_time": 0.0066, | |
| "eval_runtime": 1239.6344, | |
| "eval_samples_per_second": 87.432, | |
| "eval_steps_per_second": 2.732, | |
| "step": 7623 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7623, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.781059536914691e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |