trainer_state.json

{
  "best_metric": 0.3834666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0351/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": 0.0004813069041765065,
      "loss": 3.4628,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1736,
      "eval_loss": 3.1108431816101074,
      "eval_runtime": 5.4551,
      "eval_samples_per_second": 687.43,
      "eval_steps_per_second": 10.816,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.054651260375977,
      "learning_rate": 0.00042740099352256953,
      "loss": 2.957,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.23173333333333335,
      "eval_loss": 2.876178741455078,
      "eval_runtime": 5.7749,
      "eval_samples_per_second": 649.366,
      "eval_steps_per_second": 10.217,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.430767059326172,
      "learning_rate": 0.0003464873897488322,
      "loss": 2.7642,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2656,
      "eval_loss": 2.7746121883392334,
      "eval_runtime": 5.3618,
      "eval_samples_per_second": 699.398,
      "eval_steps_per_second": 11.004,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.488872528076172,
      "learning_rate": 0.00025088445554620134,
      "loss": 2.5823,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.312,
      "eval_loss": 2.592540740966797,
      "eval_runtime": 5.3696,
      "eval_samples_per_second": 698.382,
      "eval_steps_per_second": 10.988,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.117351531982422,
      "learning_rate": 0.00015514687100427082,
      "loss": 2.3785,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.34373333333333334,
      "eval_loss": 2.4236955642700195,
      "eval_runtime": 5.355,
      "eval_samples_per_second": 700.285,
      "eval_steps_per_second": 11.018,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.439102172851562,
      "learning_rate": 7.384981550618446e-05,
      "loss": 2.1409,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3736,
      "eval_loss": 2.331709146499634,
      "eval_runtime": 5.3598,
      "eval_samples_per_second": 699.653,
      "eval_steps_per_second": 11.008,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 20.638731002807617,
      "learning_rate": 1.9370028791882848e-05,
      "loss": 1.9211,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.38133333333333336,
      "eval_loss": 2.306528091430664,
      "eval_runtime": 5.4047,
      "eval_samples_per_second": 693.843,
      "eval_steps_per_second": 10.916,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 19.374277114868164,
      "learning_rate": 1.5645281219167639e-09,
      "loss": 1.761,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3834666666666667,
      "eval_loss": 2.3222451210021973,
      "eval_runtime": 5.3985,
      "eval_samples_per_second": 694.636,
      "eval_steps_per_second": 10.929,
      "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
}