trainer_state.json

{
  "best_metric": 0.8989333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0294/checkpoints/checkpoint-999",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 999,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.4106061458587646,
      "learning_rate": 5e-05,
      "loss": 0.7085,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.38716012239456177,
      "eval_runtime": 6.9565,
      "eval_samples_per_second": 539.061,
      "eval_steps_per_second": 8.481,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.1912,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8749333333333333,
      "eval_loss": 0.4343336820602417,
      "eval_runtime": 6.1328,
      "eval_samples_per_second": 611.465,
      "eval_steps_per_second": 9.62,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.321000099182129,
      "learning_rate": 5e-05,
      "loss": 0.1036,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.37430354952812195,
      "eval_runtime": 5.6768,
      "eval_samples_per_second": 660.589,
      "eval_steps_per_second": 10.393,
      "step": 999
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.94223961867776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}