trainer_state.json

{
  "best_metric": 0.9410666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0763/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": 5.956546783447266,
      "learning_rate": 2.401801801801802e-05,
      "loss": 0.7816,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.3686603009700775,
      "eval_runtime": 14.3948,
      "eval_samples_per_second": 260.511,
      "eval_steps_per_second": 4.099,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.055843815207481384,
      "learning_rate": 1.801801801801802e-05,
      "loss": 0.1763,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9194666666666667,
      "eval_loss": 0.26527664065361023,
      "eval_runtime": 14.0206,
      "eval_samples_per_second": 267.464,
      "eval_steps_per_second": 4.208,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 49.650508880615234,
      "learning_rate": 1.2018018018018019e-05,
      "loss": 0.0673,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9258666666666666,
      "eval_loss": 0.25297731161117554,
      "eval_runtime": 13.733,
      "eval_samples_per_second": 273.065,
      "eval_steps_per_second": 4.296,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.32931193709373474,
      "learning_rate": 6.018018018018018e-06,
      "loss": 0.0287,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.23469586670398712,
      "eval_runtime": 13.613,
      "eval_samples_per_second": 275.471,
      "eval_steps_per_second": 4.334,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6084650158882141,
      "learning_rate": 1.8018018018018017e-08,
      "loss": 0.008,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9410666666666667,
      "eval_loss": 0.21532905101776123,
      "eval_runtime": 13.461,
      "eval_samples_per_second": 278.583,
      "eval_steps_per_second": 4.383,
      "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
}