{ "best_metric": 1.0, "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-0-4-29Nov24-003/checkpoint-80", "epoch": 36.0, "eval_steps": 10, "global_step": 90, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 4.0, "grad_norm": 34.317447662353516, "learning_rate": 1e-05, "loss": 1.679, "step": 10 }, { "epoch": 4.0, "eval_accuracy": 0.5660714285714286, "eval_loss": 1.3216605186462402, "eval_runtime": 7.8392, "eval_samples_per_second": 71.436, "eval_steps_per_second": 8.929, "step": 10 }, { "epoch": 8.0, "grad_norm": 32.69915008544922, "learning_rate": 8.765432098765432e-06, "loss": 1.0782, "step": 20 }, { "epoch": 8.0, "eval_accuracy": 0.8053571428571429, "eval_loss": 0.7879157066345215, "eval_runtime": 7.7537, "eval_samples_per_second": 72.223, "eval_steps_per_second": 9.028, "step": 20 }, { "epoch": 12.0, "grad_norm": 19.49567413330078, "learning_rate": 7.530864197530865e-06, "loss": 0.6196, "step": 30 }, { "epoch": 12.0, "eval_accuracy": 0.9232142857142858, "eval_loss": 0.42588892579078674, "eval_runtime": 7.6733, "eval_samples_per_second": 72.98, "eval_steps_per_second": 9.123, "step": 30 }, { "epoch": 16.0, "grad_norm": 19.529306411743164, "learning_rate": 6.296296296296297e-06, "loss": 0.3697, "step": 40 }, { "epoch": 16.0, "eval_accuracy": 0.9553571428571429, "eval_loss": 0.2647091746330261, "eval_runtime": 7.9781, "eval_samples_per_second": 70.192, "eval_steps_per_second": 8.774, "step": 40 }, { "epoch": 20.0, "grad_norm": 11.466094970703125, "learning_rate": 5.061728395061729e-06, "loss": 0.2093, "step": 50 }, { "epoch": 20.0, "eval_accuracy": 0.9803571428571428, "eval_loss": 0.16689102351665497, "eval_runtime": 8.1259, "eval_samples_per_second": 68.915, "eval_steps_per_second": 8.614, "step": 50 }, { "epoch": 24.0, "grad_norm": 16.115257263183594, "learning_rate": 3.827160493827161e-06, "loss": 0.1362, "step": 60 }, { "epoch": 24.0, "eval_accuracy": 0.9875, "eval_loss": 0.11409498006105423, "eval_runtime": 7.6046, "eval_samples_per_second": 73.64, "eval_steps_per_second": 9.205, "step": 60 }, { "epoch": 28.0, "grad_norm": 8.264847755432129, "learning_rate": 2.5925925925925925e-06, "loss": 0.1049, "step": 70 }, { "epoch": 28.0, "eval_accuracy": 0.9928571428571429, "eval_loss": 0.09371346235275269, "eval_runtime": 8.4549, "eval_samples_per_second": 66.234, "eval_steps_per_second": 8.279, "step": 70 }, { "epoch": 32.0, "grad_norm": 12.916861534118652, "learning_rate": 1.3580246913580248e-06, "loss": 0.0809, "step": 80 }, { "epoch": 32.0, "eval_accuracy": 1.0, "eval_loss": 0.07732328772544861, "eval_runtime": 7.3493, "eval_samples_per_second": 76.198, "eval_steps_per_second": 9.525, "step": 80 }, { "epoch": 36.0, "grad_norm": 5.014369487762451, "learning_rate": 2.469135802469136e-07, "loss": 0.0731, "step": 90 }, { "epoch": 36.0, "eval_accuracy": 1.0, "eval_loss": 0.07128783315420151, "eval_runtime": 8.1342, "eval_samples_per_second": 68.846, "eval_steps_per_second": 8.606, "step": 90 }, { "epoch": 36.0, "step": 90, "total_flos": 7.889384364657869e+17, "train_loss": 0.4834257894092136, "train_runtime": 564.5005, "train_samples_per_second": 6.377, "train_steps_per_second": 0.159 } ], "logging_steps": 10, "max_steps": 90, "num_input_tokens_seen": 0, "num_train_epochs": 45, "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": 7.889384364657869e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }