{
  "best_metric": 0.41146666666666665,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0095/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": 18.986953735351562,
      "learning_rate": 0.0004755337805875084,
      "loss": 3.38,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.168,
      "eval_loss": 3.1167004108428955,
      "eval_runtime": 6.6337,
      "eval_samples_per_second": 565.299,
      "eval_steps_per_second": 8.894,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 18.289037704467773,
      "learning_rate": 0.00040639873814655834,
      "loss": 2.8719,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.25306666666666666,
      "eval_loss": 2.7620275020599365,
      "eval_runtime": 6.4865,
      "eval_samples_per_second": 578.125,
      "eval_steps_per_second": 9.096,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.234516143798828,
      "learning_rate": 0.00030628700751103485,
      "loss": 2.6345,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3,
      "eval_loss": 2.5450944900512695,
      "eval_runtime": 7.2374,
      "eval_samples_per_second": 518.14,
      "eval_steps_per_second": 8.152,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.98984718322754,
      "learning_rate": 0.00019502694472310563,
      "loss": 2.4016,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.34346666666666664,
      "eval_loss": 2.3797500133514404,
      "eval_runtime": 6.5997,
      "eval_samples_per_second": 568.209,
      "eval_steps_per_second": 8.94,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.032434463500977,
      "learning_rate": 9.465496973304394e-05,
      "loss": 2.1689,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3722666666666667,
      "eval_loss": 2.2725586891174316,
      "eval_runtime": 6.4593,
      "eval_samples_per_second": 580.554,
      "eval_steps_per_second": 9.134,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 31.766937255859375,
      "learning_rate": 2.505098316912138e-05,
      "loss": 1.9211,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4072,
      "eval_loss": 2.15832257270813,
      "eval_runtime": 6.3831,
      "eval_samples_per_second": 587.493,
      "eval_steps_per_second": 9.243,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 2.0434646496314548e-09,
      "loss": 1.7258,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.41146666666666665,
      "eval_loss": 2.1818292140960693,
      "eval_runtime": 6.6416,
      "eval_samples_per_second": 564.626,
      "eval_steps_per_second": 8.883,
      "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
}