{
  "best_metric": 0.9384,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0127/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": 6.790597915649414,
      "learning_rate": 2.572715572715573e-05,
      "loss": 0.7729,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.3210218846797943,
      "eval_runtime": 5.8834,
      "eval_samples_per_second": 637.386,
      "eval_steps_per_second": 10.028,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.1454311454311453e-05,
      "loss": 0.158,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9112,
      "eval_loss": 0.30169492959976196,
      "eval_runtime": 6.4721,
      "eval_samples_per_second": 579.409,
      "eval_steps_per_second": 9.116,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4985216856002808,
      "learning_rate": 1.716859716859717e-05,
      "loss": 0.0521,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9192,
      "eval_loss": 0.28912538290023804,
      "eval_runtime": 6.3576,
      "eval_samples_per_second": 589.848,
      "eval_steps_per_second": 9.28,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.17432601749897003,
      "learning_rate": 1.2882882882882882e-05,
      "loss": 0.0178,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.2676466703414917,
      "eval_runtime": 5.9053,
      "eval_samples_per_second": 635.028,
      "eval_steps_per_second": 9.991,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0137614905834198,
      "learning_rate": 8.597168597168598e-06,
      "loss": 0.0066,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.2661159634590149,
      "eval_runtime": 6.008,
      "eval_samples_per_second": 624.165,
      "eval_steps_per_second": 9.82,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07023172825574875,
      "learning_rate": 4.311454311454311e-06,
      "loss": 0.0013,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9365333333333333,
      "eval_loss": 0.27028369903564453,
      "eval_runtime": 5.9879,
      "eval_samples_per_second": 626.265,
      "eval_steps_per_second": 9.853,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0024981729220598936,
      "learning_rate": 2.5740025740025742e-08,
      "loss": 0.0006,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9384,
      "eval_loss": 0.2591681182384491,
      "eval_runtime": 5.9259,
      "eval_samples_per_second": 632.812,
      "eval_steps_per_second": 9.956,
      "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
}