{
  "best_metric": 0.9125333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0857/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": 4.121883392333984,
      "learning_rate": 8.728616793536588e-05,
      "loss": 1.4543,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8125333333333333,
      "eval_loss": 0.6511214971542358,
      "eval_runtime": 5.8605,
      "eval_samples_per_second": 639.874,
      "eval_steps_per_second": 10.067,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.95023019690139e-05,
      "loss": 0.4469,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.49497541785240173,
      "eval_runtime": 5.9721,
      "eval_samples_per_second": 627.921,
      "eval_steps_per_second": 9.879,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.7225565910339355,
      "learning_rate": 6.75408389678355e-05,
      "loss": 0.2575,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.4211729168891907,
      "eval_runtime": 6.3098,
      "eval_samples_per_second": 594.314,
      "eval_steps_per_second": 9.351,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1738544702529907,
      "learning_rate": 5.286061811978304e-05,
      "loss": 0.1486,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.432212769985199,
      "eval_runtime": 6.5674,
      "eval_samples_per_second": 571.005,
      "eval_steps_per_second": 8.984,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.666933536529541,
      "learning_rate": 3.723229071611673e-05,
      "loss": 0.0788,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.3941194713115692,
      "eval_runtime": 5.9093,
      "eval_samples_per_second": 634.591,
      "eval_steps_per_second": 9.984,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.24445493519306183,
      "learning_rate": 2.2540863691266368e-05,
      "loss": 0.0387,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.37708067893981934,
      "eval_runtime": 5.8748,
      "eval_samples_per_second": 638.321,
      "eval_steps_per_second": 10.043,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.675011157989502,
      "learning_rate": 1.055833996679956e-05,
      "loss": 0.0145,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.3621286451816559,
      "eval_runtime": 5.7904,
      "eval_samples_per_second": 647.629,
      "eval_steps_per_second": 10.189,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04958236217498779,
      "learning_rate": 2.7299887471028354e-06,
      "loss": 0.0071,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9104,
      "eval_loss": 0.3539009094238281,
      "eval_runtime": 5.7815,
      "eval_samples_per_second": 648.617,
      "eval_steps_per_second": 10.205,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.049867577850818634,
      "learning_rate": 2.472343088255613e-11,
      "loss": 0.0049,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.34635964035987854,
      "eval_runtime": 5.779,
      "eval_samples_per_second": 648.902,
      "eval_steps_per_second": 10.209,
      "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
}