{
  "best_metric": 0.5170666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0911/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 0.0005,
      "loss": 3.1767,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23946666666666666,
      "eval_loss": 2.8042452335357666,
      "eval_runtime": 6.4644,
      "eval_samples_per_second": 580.096,
      "eval_steps_per_second": 9.127,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 10.584012985229492,
      "learning_rate": 0.0005,
      "loss": 2.5565,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.30533333333333335,
      "eval_loss": 2.5334911346435547,
      "eval_runtime": 7.2465,
      "eval_samples_per_second": 517.489,
      "eval_steps_per_second": 8.142,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.1995,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3824,
      "eval_loss": 2.233227014541626,
      "eval_runtime": 6.4564,
      "eval_samples_per_second": 580.815,
      "eval_steps_per_second": 9.138,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.601795196533203,
      "learning_rate": 0.0005,
      "loss": 1.9219,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.43093333333333333,
      "eval_loss": 2.0613601207733154,
      "eval_runtime": 6.7566,
      "eval_samples_per_second": 555.014,
      "eval_steps_per_second": 8.732,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.5794677734375,
      "learning_rate": 0.0005,
      "loss": 1.7045,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.45013333333333333,
      "eval_loss": 1.9462791681289673,
      "eval_runtime": 7.085,
      "eval_samples_per_second": 529.287,
      "eval_steps_per_second": 8.327,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.829448699951172,
      "learning_rate": 0.0005,
      "loss": 1.4887,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4962666666666667,
      "eval_loss": 1.8062559366226196,
      "eval_runtime": 6.4421,
      "eval_samples_per_second": 582.106,
      "eval_steps_per_second": 9.158,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 12.304649353027344,
      "learning_rate": 0.0005,
      "loss": 1.2851,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5170666666666667,
      "eval_loss": 1.7313557863235474,
      "eval_runtime": 6.6901,
      "eval_samples_per_second": 560.527,
      "eval_steps_per_second": 8.819,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}