| { | |
| "best_metric": 0.8181275725364685, | |
| "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": 17.779321670532227, | |
| "learning_rate": 6.584048593688102e-07, | |
| "loss": 0.7478, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39354584809130266, | |
| "grad_norm": 19.093740463256836, | |
| "learning_rate": 6.12188036445266e-07, | |
| "loss": 0.7451, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5903187721369539, | |
| "grad_norm": 11.988561630249023, | |
| "learning_rate": 5.659712135217218e-07, | |
| "loss": 0.7481, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7870916961826053, | |
| "grad_norm": 12.481701850891113, | |
| "learning_rate": 5.197543905981778e-07, | |
| "loss": 0.7618, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9838646202282566, | |
| "grad_norm": 13.344101905822754, | |
| "learning_rate": 4.7353756767463353e-07, | |
| "loss": 0.7418, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7666260702686744, | |
| "eval_loss": 0.8249521851539612, | |
| "eval_model_preparation_time": 0.0075, | |
| "eval_runtime": 1189.4277, | |
| "eval_samples_per_second": 91.123, | |
| "eval_steps_per_second": 2.848, | |
| "step": 2541 | |
| }, | |
| { | |
| "epoch": 1.1806375442739079, | |
| "grad_norm": 12.604414939880371, | |
| "learning_rate": 4.273207447510894e-07, | |
| "loss": 0.7428, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3774104683195592, | |
| "grad_norm": 11.070133209228516, | |
| "learning_rate": 3.8110392182754523e-07, | |
| "loss": 0.7337, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5741833923652107, | |
| "grad_norm": 9.148447036743164, | |
| "learning_rate": 3.34887098904001e-07, | |
| "loss": 0.7423, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.770956316410862, | |
| "grad_norm": 8.443017959594727, | |
| "learning_rate": 2.886702759804569e-07, | |
| "loss": 0.7418, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9677292404565132, | |
| "grad_norm": 12.822120666503906, | |
| "learning_rate": 2.424534530569127e-07, | |
| "loss": 0.7395, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7675948479480366, | |
| "eval_loss": 0.819111168384552, | |
| "eval_model_preparation_time": 0.0075, | |
| "eval_runtime": 1194.3948, | |
| "eval_samples_per_second": 90.744, | |
| "eval_steps_per_second": 2.836, | |
| "step": 5082 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 8.802216529846191, | |
| "learning_rate": 1.9623663013336853e-07, | |
| "loss": 0.7385, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3612750885478158, | |
| "grad_norm": 11.116040229797363, | |
| "learning_rate": 1.5001980720982439e-07, | |
| "loss": 0.7243, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.558048012593467, | |
| "grad_norm": 8.803569793701172, | |
| "learning_rate": 1.0380298428628021e-07, | |
| "loss": 0.7448, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7548209366391183, | |
| "grad_norm": 11.275320053100586, | |
| "learning_rate": 5.7586161362736035e-08, | |
| "loss": 0.7354, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9515938606847696, | |
| "grad_norm": 12.109325408935547, | |
| "learning_rate": 1.1369338439191866e-08, | |
| "loss": 0.7297, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7687481547091821, | |
| "eval_loss": 0.8181275725364685, | |
| "eval_model_preparation_time": 0.0075, | |
| "eval_runtime": 1213.6961, | |
| "eval_samples_per_second": 89.301, | |
| "eval_steps_per_second": 2.791, | |
| "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 | |
| } | |