trainer_state.json

{
  "best_metric": 0.9186666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0807/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": 13.413515090942383,
      "learning_rate": 4.528073362367401e-05,
      "loss": 0.8011,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.5084623694419861,
      "eval_runtime": 10.7738,
      "eval_samples_per_second": 348.065,
      "eval_steps_per_second": 5.476,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.02672097086906433,
      "learning_rate": 3.281509432839476e-05,
      "loss": 0.2075,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.42071008682250977,
      "eval_runtime": 10.9288,
      "eval_samples_per_second": 343.13,
      "eval_steps_per_second": 5.399,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 37.35453414916992,
      "learning_rate": 1.736435462495525e-05,
      "loss": 0.0685,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.36948758363723755,
      "eval_runtime": 11.0812,
      "eval_samples_per_second": 338.412,
      "eval_steps_per_second": 5.324,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6655235886573792,
      "learning_rate": 4.8301719287419e-06,
      "loss": 0.0152,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.32354485988616943,
      "eval_runtime": 10.9082,
      "eval_samples_per_second": 343.777,
      "eval_steps_per_second": 5.409,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.21489106118679047,
      "learning_rate": 1.7800850740101738e-10,
      "loss": 0.0029,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9186666666666666,
      "eval_loss": 0.3130680322647095,
      "eval_runtime": 10.6801,
      "eval_samples_per_second": 351.122,
      "eval_steps_per_second": 5.524,
      "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
}