{ "best_metric": 0.9732142857142857, "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-0-4-30Nov24-008/checkpoint-50", "epoch": 31.90909090909091, "eval_steps": 10, "global_step": 80, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 3.909090909090909, "grad_norm": 22.989723205566406, "learning_rate": 9.861111111111112e-06, "loss": 1.538, "step": 10 }, { "epoch": 3.909090909090909, "eval_accuracy": 0.5982142857142857, "eval_loss": 1.2297297716140747, "eval_runtime": 7.4154, "eval_samples_per_second": 75.518, "eval_steps_per_second": 9.44, "step": 10 }, { "epoch": 7.909090909090909, "grad_norm": 16.446542739868164, "learning_rate": 8.472222222222223e-06, "loss": 0.7675, "step": 20 }, { "epoch": 7.909090909090909, "eval_accuracy": 0.9321428571428572, "eval_loss": 0.5667614340782166, "eval_runtime": 8.6379, "eval_samples_per_second": 64.831, "eval_steps_per_second": 8.104, "step": 20 }, { "epoch": 11.909090909090908, "grad_norm": 10.684986114501953, "learning_rate": 7.083333333333335e-06, "loss": 0.2383, "step": 30 }, { "epoch": 11.909090909090908, "eval_accuracy": 0.95, "eval_loss": 0.273371160030365, "eval_runtime": 8.248, "eval_samples_per_second": 67.895, "eval_steps_per_second": 8.487, "step": 30 }, { "epoch": 15.909090909090908, "grad_norm": 2.3718950748443604, "learning_rate": 5.694444444444445e-06, "loss": 0.0619, "step": 40 }, { "epoch": 15.909090909090908, "eval_accuracy": 0.9642857142857143, "eval_loss": 0.15701523423194885, "eval_runtime": 8.3108, "eval_samples_per_second": 67.382, "eval_steps_per_second": 8.423, "step": 40 }, { "epoch": 19.90909090909091, "grad_norm": 0.8196462988853455, "learning_rate": 4.305555555555556e-06, "loss": 0.0172, "step": 50 }, { "epoch": 19.90909090909091, "eval_accuracy": 0.9732142857142857, "eval_loss": 0.10191339254379272, "eval_runtime": 9.1466, "eval_samples_per_second": 61.225, "eval_steps_per_second": 7.653, "step": 50 }, { "epoch": 23.90909090909091, "grad_norm": 0.3251395523548126, "learning_rate": 2.916666666666667e-06, "loss": 0.0058, "step": 60 }, { "epoch": 23.90909090909091, "eval_accuracy": 0.9696428571428571, "eval_loss": 0.09441839158535004, "eval_runtime": 8.7407, "eval_samples_per_second": 64.068, "eval_steps_per_second": 8.008, "step": 60 }, { "epoch": 27.90909090909091, "grad_norm": 0.20590414106845856, "learning_rate": 1.527777777777778e-06, "loss": 0.0031, "step": 70 }, { "epoch": 27.90909090909091, "eval_accuracy": 0.9714285714285714, "eval_loss": 0.08477898687124252, "eval_runtime": 8.7431, "eval_samples_per_second": 64.05, "eval_steps_per_second": 8.006, "step": 70 }, { "epoch": 31.90909090909091, "grad_norm": 0.13806365430355072, "learning_rate": 1.3888888888888888e-07, "loss": 0.0024, "step": 80 }, { "epoch": 31.90909090909091, "eval_accuracy": 0.9714285714285714, "eval_loss": 0.08295369893312454, "eval_runtime": 9.0216, "eval_samples_per_second": 62.073, "eval_steps_per_second": 7.759, "step": 80 }, { "epoch": 31.90909090909091, "step": 80, "total_flos": 7.012786101918106e+17, "train_loss": 0.32927216980606316, "train_runtime": 524.626, "train_samples_per_second": 6.252, "train_steps_per_second": 0.152 } ], "logging_steps": 10, "max_steps": 80, "num_input_tokens_seen": 0, "num_train_epochs": 40, "save_steps": 10, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 7.012786101918106e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }