trainer_state.json

{
  "best_metric": 0.9213333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0998/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9979705810546875,
      "learning_rate": 9.700253128109275e-05,
      "loss": 0.8698,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8154666666666667,
      "eval_loss": 0.5987895727157593,
      "eval_runtime": 11.1476,
      "eval_samples_per_second": 336.395,
      "eval_steps_per_second": 5.293,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 64.08434295654297,
      "learning_rate": 8.836951787320407e-05,
      "loss": 0.3287,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.5621266961097717,
      "eval_runtime": 10.5055,
      "eval_samples_per_second": 356.954,
      "eval_steps_per_second": 5.616,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.320358276367188,
      "learning_rate": 7.509072574151719e-05,
      "loss": 0.1863,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8621333333333333,
      "eval_loss": 0.507114052772522,
      "eval_runtime": 10.5244,
      "eval_samples_per_second": 356.313,
      "eval_steps_per_second": 5.606,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.18623174726963043,
      "learning_rate": 5.8785621785729404e-05,
      "loss": 0.1054,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.49387046694755554,
      "eval_runtime": 10.8791,
      "eval_samples_per_second": 344.697,
      "eval_steps_per_second": 5.423,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.02863243967294693,
      "learning_rate": 4.142084218061449e-05,
      "loss": 0.0515,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.4524930417537689,
      "eval_runtime": 10.6229,
      "eval_samples_per_second": 353.01,
      "eval_steps_per_second": 5.554,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.041576988995075226,
      "learning_rate": 2.5090835623402032e-05,
      "loss": 0.0172,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9072,
      "eval_loss": 0.3993571400642395,
      "eval_runtime": 10.6798,
      "eval_samples_per_second": 351.131,
      "eval_steps_per_second": 5.524,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0018848832696676254,
      "learning_rate": 1.1765241910120828e-05,
      "loss": 0.0044,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.35042083263397217,
      "eval_runtime": 10.6517,
      "eval_samples_per_second": 352.056,
      "eval_steps_per_second": 5.539,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0007570235757157207,
      "learning_rate": 3.0513243074104303e-06,
      "loss": 0.0014,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9213333333333333,
      "eval_loss": 0.33495256304740906,
      "eval_runtime": 10.4242,
      "eval_samples_per_second": 359.741,
      "eval_steps_per_second": 5.66,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}