{
  "best_metric": 0.9477333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0184/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.6929118633270264,
      "learning_rate": 8.75e-05,
      "loss": 0.5972,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.24695582687854767,
      "eval_runtime": 6.3822,
      "eval_samples_per_second": 587.573,
      "eval_steps_per_second": 9.244,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.10167285799980164,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1196,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.25699925422668457,
      "eval_runtime": 6.532,
      "eval_samples_per_second": 574.093,
      "eval_steps_per_second": 9.032,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.020267032086849213,
      "learning_rate": 6.25e-05,
      "loss": 0.0499,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9325333333333333,
      "eval_loss": 0.2473408579826355,
      "eval_runtime": 6.1026,
      "eval_samples_per_second": 614.489,
      "eval_steps_per_second": 9.668,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.008303971961140633,
      "learning_rate": 5e-05,
      "loss": 0.0171,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9421333333333334,
      "eval_loss": 0.23549091815948486,
      "eval_runtime": 6.1802,
      "eval_samples_per_second": 606.773,
      "eval_steps_per_second": 9.547,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.009643266908824444,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.005,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9392,
      "eval_loss": 0.2537183165550232,
      "eval_runtime": 6.6758,
      "eval_samples_per_second": 561.727,
      "eval_steps_per_second": 8.838,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.024560758844017982,
      "learning_rate": 2.5e-05,
      "loss": 0.0022,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9456,
      "eval_loss": 0.23798252642154694,
      "eval_runtime": 6.5515,
      "eval_samples_per_second": 572.389,
      "eval_steps_per_second": 9.006,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 1.2537537537537538e-05,
      "loss": 0.001,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9474666666666667,
      "eval_loss": 0.23667828738689423,
      "eval_runtime": 7.3378,
      "eval_samples_per_second": 511.053,
      "eval_steps_per_second": 8.041,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0052330526523292065,
      "learning_rate": 3.753753753753754e-08,
      "loss": 0.0007,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9477333333333333,
      "eval_loss": 0.23786941170692444,
      "eval_runtime": 6.0391,
      "eval_samples_per_second": 620.951,
      "eval_steps_per_second": 9.77,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}