trainer_state.json

{
  "best_metric": 0.8432,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0747/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": 1.6839598417282104,
      "learning_rate": 0.0001,
      "loss": 0.9074,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7856,
      "eval_loss": 0.7049508690834045,
      "eval_runtime": 11.099,
      "eval_samples_per_second": 337.868,
      "eval_steps_per_second": 5.316,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8743887543678284,
      "learning_rate": 0.0001,
      "loss": 0.3911,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8285333333333333,
      "eval_loss": 0.5812381505966187,
      "eval_runtime": 11.0629,
      "eval_samples_per_second": 338.971,
      "eval_steps_per_second": 5.333,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 150.55953979492188,
      "learning_rate": 0.0001,
      "loss": 0.2677,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8397333333333333,
      "eval_loss": 0.5773179531097412,
      "eval_runtime": 11.0106,
      "eval_samples_per_second": 340.582,
      "eval_steps_per_second": 5.358,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9587396383285522,
      "learning_rate": 0.0001,
      "loss": 0.2046,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8266666666666667,
      "eval_loss": 0.6550912261009216,
      "eval_runtime": 10.6706,
      "eval_samples_per_second": 351.432,
      "eval_steps_per_second": 5.529,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.431104898452759,
      "learning_rate": 0.0001,
      "loss": 0.1654,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8432,
      "eval_loss": 0.6395440101623535,
      "eval_runtime": 10.6453,
      "eval_samples_per_second": 352.268,
      "eval_steps_per_second": 5.542,
      "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
}