{
  "best_metric": 0.8744,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0882/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": 156.16111755371094,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.7875,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.728,
      "eval_loss": 1.0363245010375977,
      "eval_runtime": 7.9623,
      "eval_samples_per_second": 470.971,
      "eval_steps_per_second": 7.41,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 122.00597381591797,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7448,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8248,
      "eval_loss": 0.5964208245277405,
      "eval_runtime": 7.5258,
      "eval_samples_per_second": 498.288,
      "eval_steps_per_second": 7.84,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 64.17169952392578,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.4703,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.48978209495544434,
      "eval_runtime": 7.4684,
      "eval_samples_per_second": 502.113,
      "eval_steps_per_second": 7.9,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 87.70691680908203,
      "learning_rate": 5e-05,
      "loss": 0.3561,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.45913729071617126,
      "eval_runtime": 7.4115,
      "eval_samples_per_second": 505.968,
      "eval_steps_per_second": 7.961,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 88.71204376220703,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.2841,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.43221867084503174,
      "eval_runtime": 7.4315,
      "eval_samples_per_second": 504.609,
      "eval_steps_per_second": 7.939,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 125.51765441894531,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.2417,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.43772247433662415,
      "eval_runtime": 7.422,
      "eval_samples_per_second": 505.257,
      "eval_steps_per_second": 7.949,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 43.24335479736328,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.2211,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.42791983485221863,
      "eval_runtime": 7.4575,
      "eval_samples_per_second": 502.852,
      "eval_steps_per_second": 7.912,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}