trainer_state.json

{
  "best_metric": 0.9173333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0214/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": 2.9106113962682575e-05,
      "loss": 0.8864,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.3911151885986328,
      "eval_runtime": 7.5834,
      "eval_samples_per_second": 494.499,
      "eval_steps_per_second": 7.78,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.862834095954895,
      "learning_rate": 2.6510855361961218e-05,
      "loss": 0.2574,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.3737613260746002,
      "eval_runtime": 6.8278,
      "eval_samples_per_second": 549.224,
      "eval_steps_per_second": 8.641,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7440066337585449,
      "learning_rate": 2.2527217722455156e-05,
      "loss": 0.122,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.3564225137233734,
      "eval_runtime": 7.1896,
      "eval_samples_per_second": 521.586,
      "eval_steps_per_second": 8.206,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.841376543045044,
      "learning_rate": 1.763568653571882e-05,
      "loss": 0.0608,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.3442498743534088,
      "eval_runtime": 6.7858,
      "eval_samples_per_second": 552.623,
      "eval_steps_per_second": 8.695,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.614599585533142,
      "learning_rate": 1.2426252654184347e-05,
      "loss": 0.028,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9034666666666666,
      "eval_loss": 0.3722509741783142,
      "eval_runtime": 6.9293,
      "eval_samples_per_second": 541.182,
      "eval_steps_per_second": 8.515,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0018188483081758022,
      "learning_rate": 7.5272506870206094e-06,
      "loss": 0.0112,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9136,
      "eval_loss": 0.3522760272026062,
      "eval_runtime": 6.942,
      "eval_samples_per_second": 540.188,
      "eval_steps_per_second": 8.499,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.02974710986018181,
      "learning_rate": 3.5295725730362483e-06,
      "loss": 0.0038,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9141333333333334,
      "eval_loss": 0.3562905192375183,
      "eval_runtime": 6.6936,
      "eval_samples_per_second": 560.238,
      "eval_steps_per_second": 8.814,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.008892694488167763,
      "learning_rate": 9.153972922231291e-07,
      "loss": 0.0014,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3545072674751282,
      "eval_runtime": 7.3253,
      "eval_samples_per_second": 511.923,
      "eval_steps_per_second": 8.054,
      "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
}