trainer_state.json

{
  "best_metric": 0.23813333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0270/checkpoints/checkpoint-666",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 666,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 24.095142364501953,
      "learning_rate": 0.0003,
      "loss": 3.3589,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18933333333333333,
      "eval_loss": 3.0369999408721924,
      "eval_runtime": 6.9486,
      "eval_samples_per_second": 539.675,
      "eval_steps_per_second": 8.491,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 16.27484130859375,
      "learning_rate": 0.0003,
      "loss": 2.8134,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.23813333333333334,
      "eval_loss": 2.8394808769226074,
      "eval_runtime": 7.3703,
      "eval_samples_per_second": 508.801,
      "eval_steps_per_second": 8.005,
      "step": 666
    }
  ],
  "logging_steps": 500,
  "max_steps": 666,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 3.29482641245184e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}