{
  "best_metric": 0.8557333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0367/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": 103.17560577392578,
      "learning_rate": 6.653391037658466e-05,
      "loss": 3.4102,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5634666666666667,
      "eval_loss": 1.9522404670715332,
      "eval_runtime": 13.1551,
      "eval_samples_per_second": 285.061,
      "eval_steps_per_second": 4.485,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 131.11557006835938,
      "learning_rate": 5.682214306505567e-05,
      "loss": 1.144,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7970666666666667,
      "eval_loss": 0.7198318243026733,
      "eval_runtime": 13.2252,
      "eval_samples_per_second": 283.549,
      "eval_steps_per_second": 4.461,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 91.01720428466797,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.6579,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8341333333333333,
      "eval_loss": 0.5647615790367126,
      "eval_runtime": 13.1647,
      "eval_samples_per_second": 284.853,
      "eval_steps_per_second": 4.482,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 85.84105682373047,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.52,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.49526938796043396,
      "eval_runtime": 13.2338,
      "eval_samples_per_second": 283.366,
      "eval_steps_per_second": 4.458,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 127.80591583251953,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.455,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.49106231331825256,
      "eval_runtime": 13.1668,
      "eval_samples_per_second": 284.806,
      "eval_steps_per_second": 4.481,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 71.15036010742188,
      "learning_rate": 3.4660896234153336e-06,
      "loss": 0.4087,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.46839573979377747,
      "eval_runtime": 14.4405,
      "eval_samples_per_second": 259.687,
      "eval_steps_per_second": 4.086,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 150.55902099609375,
      "learning_rate": 0.0,
      "loss": 0.3974,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8557333333333333,
      "eval_loss": 0.48331770300865173,
      "eval_runtime": 14.0288,
      "eval_samples_per_second": 267.307,
      "eval_steps_per_second": 4.206,
      "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}