{
  "best_metric": 0.9042666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0012/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 26.564544677734375,
      "learning_rate": 0.0005,
      "loss": 1.1537,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.47399577498435974,
      "eval_runtime": 8.4498,
      "eval_samples_per_second": 443.8,
      "eval_steps_per_second": 6.982,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 45.01036071777344,
      "learning_rate": 0.0005,
      "loss": 0.3781,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.41294345259666443,
      "eval_runtime": 8.305,
      "eval_samples_per_second": 451.536,
      "eval_steps_per_second": 7.104,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 32.906028747558594,
      "learning_rate": 0.0005,
      "loss": 0.2419,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4254502058029175,
      "eval_runtime": 8.3703,
      "eval_samples_per_second": 448.012,
      "eval_steps_per_second": 7.049,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 42.54781723022461,
      "learning_rate": 0.0005,
      "loss": 0.179,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.385920912027359,
      "eval_runtime": 8.6438,
      "eval_samples_per_second": 433.836,
      "eval_steps_per_second": 6.826,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 18.312402725219727,
      "learning_rate": 0.0005,
      "loss": 0.121,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.3855483829975128,
      "eval_runtime": 8.463,
      "eval_samples_per_second": 443.105,
      "eval_steps_per_second": 6.972,
      "step": 1665
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}