{ "best_metric": 0.30082234740257263, "best_model_checkpoint": "./vit-base-beans/checkpoint-40", "epoch": 3.1578947368421053, "global_step": 120, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.26, "learning_rate": 0.00018684210526315792, "loss": 0.3975, "step": 10 }, { "epoch": 0.53, "learning_rate": 0.0001736842105263158, "loss": 0.1217, "step": 20 }, { "epoch": 0.79, "learning_rate": 0.0001605263157894737, "loss": 0.0602, "step": 30 }, { "epoch": 1.05, "learning_rate": 0.00014736842105263158, "loss": 0.0518, "step": 40 }, { "epoch": 1.05, "eval_accuracy": 0.9090909090909091, "eval_loss": 0.30082234740257263, "eval_runtime": 0.4635, "eval_samples_per_second": 47.461, "eval_steps_per_second": 6.472, "step": 40 }, { "epoch": 1.32, "learning_rate": 0.00013421052631578948, "loss": 0.0332, "step": 50 }, { "epoch": 1.58, "learning_rate": 0.00012105263157894738, "loss": 0.0284, "step": 60 }, { "epoch": 1.84, "learning_rate": 0.00010789473684210527, "loss": 0.0145, "step": 70 }, { "epoch": 2.11, "learning_rate": 9.473684210526316e-05, "loss": 0.0063, "step": 80 }, { "epoch": 2.11, "eval_accuracy": 0.8636363636363636, "eval_loss": 0.4024747908115387, "eval_runtime": 0.32, "eval_samples_per_second": 68.747, "eval_steps_per_second": 9.375, "step": 80 }, { "epoch": 2.37, "learning_rate": 8.157894736842105e-05, "loss": 0.0056, "step": 90 }, { "epoch": 2.63, "learning_rate": 6.842105263157895e-05, "loss": 0.0052, "step": 100 }, { "epoch": 2.89, "learning_rate": 5.526315789473685e-05, "loss": 0.0046, "step": 110 }, { "epoch": 3.16, "learning_rate": 4.210526315789474e-05, "loss": 0.0045, "step": 120 }, { "epoch": 3.16, "eval_accuracy": 0.8636363636363636, "eval_loss": 0.3997698724269867, "eval_runtime": 0.3323, "eval_samples_per_second": 66.204, "eval_steps_per_second": 9.028, "step": 120 } ], "max_steps": 152, "num_train_epochs": 4, "total_flos": 1.4552995649620378e+17, "trial_name": null, "trial_params": null }