trainer_state.json

{
  "best_metric": 0.8186666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0378/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": 8.608138084411621,
      "learning_rate": 0.0001,
      "loss": 0.9557,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.788,
      "eval_loss": 0.6846716403961182,
      "eval_runtime": 5.2271,
      "eval_samples_per_second": 717.413,
      "eval_steps_per_second": 11.287,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3729,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8186666666666667,
      "eval_loss": 0.5941157341003418,
      "eval_runtime": 5.2992,
      "eval_samples_per_second": 707.648,
      "eval_steps_per_second": 11.134,
      "step": 666
    }
  ],
  "logging_steps": 500,
  "max_steps": 1332,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 3.29482641245184e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}