trainer_state.json

{
  "best_metric": 0.9048,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0291/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": Infinity,
      "learning_rate": 7.885135135135135e-05,
      "loss": 0.8129,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8293333333333334,
      "eval_loss": 0.5601472854614258,
      "eval_runtime": 6.7912,
      "eval_samples_per_second": 552.186,
      "eval_steps_per_second": 8.688,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 143.17764282226562,
      "learning_rate": 6.760135135135137e-05,
      "loss": 0.2705,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.470471054315567,
      "eval_runtime": 6.2914,
      "eval_samples_per_second": 596.053,
      "eval_steps_per_second": 9.378,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 69.28026580810547,
      "learning_rate": 5.635135135135136e-05,
      "loss": 0.1348,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.5373799204826355,
      "eval_runtime": 6.2061,
      "eval_samples_per_second": 604.247,
      "eval_steps_per_second": 9.507,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.007216832600533962,
      "learning_rate": 4.510135135135136e-05,
      "loss": 0.07,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.4658820629119873,
      "eval_runtime": 6.7036,
      "eval_samples_per_second": 559.404,
      "eval_steps_per_second": 8.801,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04990759864449501,
      "learning_rate": 3.385135135135135e-05,
      "loss": 0.0339,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.511195957660675,
      "eval_runtime": 6.149,
      "eval_samples_per_second": 609.856,
      "eval_steps_per_second": 9.595,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.6738038063049316,
      "learning_rate": 2.2601351351351353e-05,
      "loss": 0.0132,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.4539289176464081,
      "eval_runtime": 6.7212,
      "eval_samples_per_second": 557.934,
      "eval_steps_per_second": 8.778,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.22590245306491852,
      "learning_rate": 1.135135135135135e-05,
      "loss": 0.0029,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9013333333333333,
      "eval_loss": 0.41221192479133606,
      "eval_runtime": 6.6171,
      "eval_samples_per_second": 566.709,
      "eval_steps_per_second": 8.916,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.05093248188495636,
      "learning_rate": 1.0135135135135135e-07,
      "loss": 0.001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9048,
      "eval_loss": 0.3993247449398041,
      "eval_runtime": 6.3055,
      "eval_samples_per_second": 594.721,
      "eval_steps_per_second": 9.357,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}