trainer_state.json

{
  "best_metric": 0.9394666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0544/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": 4.811948852959657e-05,
      "loss": 0.7717,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.4106404185295105,
      "eval_runtime": 10.7863,
      "eval_samples_per_second": 347.663,
      "eval_steps_per_second": 5.47,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.31267049908638,
      "learning_rate": 4.271931404387096e-05,
      "loss": 0.1948,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.34222403168678284,
      "eval_runtime": 11.1114,
      "eval_samples_per_second": 337.493,
      "eval_steps_per_second": 5.31,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.245376586914062,
      "learning_rate": 3.4621534620947785e-05,
      "loss": 0.0899,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9104,
      "eval_loss": 0.3417433798313141,
      "eval_runtime": 11.5225,
      "eval_samples_per_second": 325.451,
      "eval_steps_per_second": 5.12,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7824826836585999,
      "learning_rate": 2.5058963771413844e-05,
      "loss": 0.0406,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9162666666666667,
      "eval_loss": 0.3153071403503418,
      "eval_runtime": 10.3598,
      "eval_samples_per_second": 361.978,
      "eval_steps_per_second": 5.695,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0011729048565030098,
      "learning_rate": 1.5487416222190068e-05,
      "loss": 0.0111,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.28869935870170593,
      "eval_runtime": 10.8782,
      "eval_samples_per_second": 344.727,
      "eval_steps_per_second": 5.424,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0010228686733171344,
      "learning_rate": 7.36407332135117e-06,
      "loss": 0.0021,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9370666666666667,
      "eval_loss": 0.28418704867362976,
      "eval_runtime": 10.5461,
      "eval_samples_per_second": 355.583,
      "eval_steps_per_second": 5.595,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0012757960939779878,
      "learning_rate": 1.925640387263111e-06,
      "loss": 0.0004,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9392,
      "eval_loss": 0.2871142029762268,
      "eval_runtime": 10.0939,
      "eval_samples_per_second": 371.512,
      "eval_steps_per_second": 5.845,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0021611531265079975,
      "learning_rate": 6.953462348691098e-11,
      "loss": 0.0001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9394666666666667,
      "eval_loss": 0.2869276702404022,
      "eval_runtime": 10.0914,
      "eval_samples_per_second": 371.604,
      "eval_steps_per_second": 5.847,
      "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
}