trainer_state.json

{
  "best_metric": 0.9141333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0689/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 77.37686920166016,
      "learning_rate": 6.533836810383375e-05,
      "loss": 0.869,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8253333333333334,
      "eval_loss": 0.5526688694953918,
      "eval_runtime": 11.2319,
      "eval_samples_per_second": 333.871,
      "eval_steps_per_second": 5.253,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5929165482521057,
      "learning_rate": 5.2547638240566195e-05,
      "loss": 0.3057,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8448,
      "eval_loss": 0.5182108283042908,
      "eval_runtime": 11.5177,
      "eval_samples_per_second": 325.585,
      "eval_steps_per_second": 5.123,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 49.62903594970703,
      "learning_rate": 3.505503288166546e-05,
      "loss": 0.1432,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.4393262267112732,
      "eval_runtime": 11.0671,
      "eval_samples_per_second": 338.842,
      "eval_steps_per_second": 5.331,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.02026299014687538,
      "learning_rate": 1.754768150656529e-05,
      "loss": 0.0558,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9045333333333333,
      "eval_loss": 0.3556371331214905,
      "eval_runtime": 10.9645,
      "eval_samples_per_second": 342.013,
      "eval_steps_per_second": 5.381,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0015936290146782994,
      "learning_rate": 4.7166647778317115e-06,
      "loss": 0.017,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.908,
      "eval_loss": 0.34440189599990845,
      "eval_runtime": 11.219,
      "eval_samples_per_second": 334.253,
      "eval_steps_per_second": 5.259,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.0040788650512695,
      "learning_rate": 4.3265999090102975e-11,
      "loss": 0.0046,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9141333333333334,
      "eval_loss": 0.33373814821243286,
      "eval_runtime": 10.9596,
      "eval_samples_per_second": 342.167,
      "eval_steps_per_second": 5.383,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}