trainer_state.json

{
  "best_metric": 0.9136,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0217/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": 19.108074188232422,
      "learning_rate": 5.6084084084084077e-05,
      "loss": 0.8547,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8285333333333333,
      "eval_loss": 0.5244670510292053,
      "eval_runtime": 7.3429,
      "eval_samples_per_second": 510.694,
      "eval_steps_per_second": 8.035,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.143407821655273,
      "learning_rate": 4.208408408408408e-05,
      "loss": 0.2837,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.47544968128204346,
      "eval_runtime": 7.6708,
      "eval_samples_per_second": 488.864,
      "eval_steps_per_second": 7.691,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06139848381280899,
      "learning_rate": 2.808408408408408e-05,
      "loss": 0.1219,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.39091387391090393,
      "eval_runtime": 7.861,
      "eval_samples_per_second": 477.037,
      "eval_steps_per_second": 7.505,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0015266473637893796,
      "learning_rate": 1.4084084084084082e-05,
      "loss": 0.0528,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.3381854295730591,
      "eval_runtime": 7.3529,
      "eval_samples_per_second": 510.006,
      "eval_steps_per_second": 8.024,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.017959004268050194,
      "learning_rate": 8.408408408408407e-08,
      "loss": 0.0155,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9136,
      "eval_loss": 0.33044421672821045,
      "eval_runtime": 8.649,
      "eval_samples_per_second": 433.576,
      "eval_steps_per_second": 6.822,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}