trainer_state.json

{
  "best_metric": 0.9226666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0691/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": 7.277786731719971,
      "learning_rate": 7.878378378378379e-05,
      "loss": 0.8632,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5186222195625305,
      "eval_runtime": 13.1877,
      "eval_samples_per_second": 284.356,
      "eval_steps_per_second": 4.474,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 6.756756756756757e-05,
      "loss": 0.3245,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4622018337249756,
      "eval_runtime": 13.2205,
      "eval_samples_per_second": 283.649,
      "eval_steps_per_second": 4.463,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.288386344909668,
      "learning_rate": 5.6317567567567566e-05,
      "loss": 0.1826,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4388185441493988,
      "eval_runtime": 12.2587,
      "eval_samples_per_second": 305.906,
      "eval_steps_per_second": 4.813,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.5911848545074463,
      "learning_rate": 4.5067567567567564e-05,
      "loss": 0.1022,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.42018553614616394,
      "eval_runtime": 12.1157,
      "eval_samples_per_second": 309.516,
      "eval_steps_per_second": 4.87,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.842790603637695,
      "learning_rate": 3.381756756756757e-05,
      "loss": 0.0578,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.4315657317638397,
      "eval_runtime": 12.5972,
      "eval_samples_per_second": 297.686,
      "eval_steps_per_second": 4.684,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.006229918450117111,
      "learning_rate": 2.256756756756757e-05,
      "loss": 0.029,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9074666666666666,
      "eval_loss": 0.3863919973373413,
      "eval_runtime": 11.8387,
      "eval_samples_per_second": 316.758,
      "eval_steps_per_second": 4.984,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0001675039966357872,
      "learning_rate": 1.1317567567567568e-05,
      "loss": 0.0101,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.36882176995277405,
      "eval_runtime": 11.6965,
      "eval_samples_per_second": 320.61,
      "eval_steps_per_second": 5.044,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.3909738063812256,
      "learning_rate": 6.756756756756757e-08,
      "loss": 0.0037,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.33963459730148315,
      "eval_runtime": 13.0264,
      "eval_samples_per_second": 287.877,
      "eval_steps_per_second": 4.529,
      "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
}