{
  "best_metric": 0.868,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0125/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 1.7341,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7970666666666667,
      "eval_loss": 0.7307370901107788,
      "eval_runtime": 6.0716,
      "eval_samples_per_second": 617.634,
      "eval_steps_per_second": 9.717,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.5469,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5637804269790649,
      "eval_runtime": 6.1356,
      "eval_samples_per_second": 611.188,
      "eval_steps_per_second": 9.616,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5474292039871216,
      "learning_rate": 7e-05,
      "loss": 0.3544,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8442666666666667,
      "eval_loss": 0.5264011025428772,
      "eval_runtime": 6.4896,
      "eval_samples_per_second": 577.845,
      "eval_steps_per_second": 9.091,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 64.49068450927734,
      "learning_rate": 7e-05,
      "loss": 0.2537,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.44891634583473206,
      "eval_runtime": 6.0344,
      "eval_samples_per_second": 621.435,
      "eval_steps_per_second": 9.777,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0308116674423218,
      "learning_rate": 7e-05,
      "loss": 0.1854,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4552401602268219,
      "eval_runtime": 6.0753,
      "eval_samples_per_second": 617.254,
      "eval_steps_per_second": 9.711,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8763220310211182,
      "learning_rate": 7e-05,
      "loss": 0.1437,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8530666666666666,
      "eval_loss": 0.4950791597366333,
      "eval_runtime": 6.5524,
      "eval_samples_per_second": 572.313,
      "eval_steps_per_second": 9.004,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.04716915264725685,
      "learning_rate": 7e-05,
      "loss": 0.1269,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4925222098827362,
      "eval_runtime": 7.5082,
      "eval_samples_per_second": 499.456,
      "eval_steps_per_second": 7.858,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.036299414932727814,
      "learning_rate": 7e-05,
      "loss": 0.105,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.5248702764511108,
      "eval_runtime": 7.1166,
      "eval_samples_per_second": 526.935,
      "eval_steps_per_second": 8.29,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.6750072836875916,
      "learning_rate": 7e-05,
      "loss": 0.0887,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.5103010535240173,
      "eval_runtime": 6.5343,
      "eval_samples_per_second": 573.898,
      "eval_steps_per_second": 9.029,
      "step": 2997
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.482671885603328e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}