{ "best_metric": 0.9625, "best_model_checkpoint": "vit-large-patch16-224-in21k-dungeon-geo-morphs-0-4-30Nov24-004/checkpoint-80", "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": 15.283150672912598, "learning_rate": 1.9722222222222224e-05, "loss": 1.3834, "step": 10 }, { "epoch": 3.909090909090909, "eval_accuracy": 0.7928571428571428, "eval_loss": 1.1055018901824951, "eval_runtime": 7.219, "eval_samples_per_second": 77.573, "eval_steps_per_second": 9.697, "step": 10 }, { "epoch": 7.909090909090909, "grad_norm": 7.206173419952393, "learning_rate": 1.6944444444444446e-05, "loss": 0.5606, "step": 20 }, { "epoch": 7.909090909090909, "eval_accuracy": 0.9285714285714286, "eval_loss": 0.5140870213508606, "eval_runtime": 7.6693, "eval_samples_per_second": 73.018, "eval_steps_per_second": 9.127, "step": 20 }, { "epoch": 11.909090909090908, "grad_norm": 3.0234339237213135, "learning_rate": 1.416666666666667e-05, "loss": 0.13, "step": 30 }, { "epoch": 11.909090909090908, "eval_accuracy": 0.9517857142857142, "eval_loss": 0.2629169523715973, "eval_runtime": 7.8377, "eval_samples_per_second": 71.45, "eval_steps_per_second": 8.931, "step": 30 }, { "epoch": 15.909090909090908, "grad_norm": 0.5464780926704407, "learning_rate": 1.138888888888889e-05, "loss": 0.0283, "step": 40 }, { "epoch": 15.909090909090908, "eval_accuracy": 0.9464285714285714, "eval_loss": 0.16540393233299255, "eval_runtime": 7.3817, "eval_samples_per_second": 75.863, "eval_steps_per_second": 9.483, "step": 40 }, { "epoch": 19.90909090909091, "grad_norm": 0.23106057941913605, "learning_rate": 8.611111111111112e-06, "loss": 0.0082, "step": 50 }, { "epoch": 19.90909090909091, "eval_accuracy": 0.9553571428571429, "eval_loss": 0.1351735144853592, "eval_runtime": 8.4845, "eval_samples_per_second": 66.003, "eval_steps_per_second": 8.25, "step": 50 }, { "epoch": 23.90909090909091, "grad_norm": 0.1367451250553131, "learning_rate": 5.833333333333334e-06, "loss": 0.0043, "step": 60 }, { "epoch": 23.90909090909091, "eval_accuracy": 0.9589285714285715, "eval_loss": 0.13371427357196808, "eval_runtime": 7.4124, "eval_samples_per_second": 75.55, "eval_steps_per_second": 9.444, "step": 60 }, { "epoch": 27.90909090909091, "grad_norm": 0.106055848300457, "learning_rate": 3.055555555555556e-06, "loss": 0.0033, "step": 70 }, { "epoch": 27.90909090909091, "eval_accuracy": 0.9607142857142857, "eval_loss": 0.12570610642433167, "eval_runtime": 7.353, "eval_samples_per_second": 76.16, "eval_steps_per_second": 9.52, "step": 70 }, { "epoch": 31.90909090909091, "grad_norm": 0.09179096668958664, "learning_rate": 2.7777777777777776e-07, "loss": 0.0029, "step": 80 }, { "epoch": 31.90909090909091, "eval_accuracy": 0.9625, "eval_loss": 0.12754002213478088, "eval_runtime": 7.2874, "eval_samples_per_second": 76.845, "eval_steps_per_second": 9.606, "step": 80 }, { "epoch": 31.90909090909091, "step": 80, "total_flos": 7.012786101918106e+17, "train_loss": 0.2651173733174801, "train_runtime": 536.5409, "train_samples_per_second": 6.113, "train_steps_per_second": 0.149 } ], "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 }