trainer_state.json

{
  "best_metric": 0.9314666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0493/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 9.06165595011943e-05,
      "loss": 0.8198,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5335952639579773,
      "eval_runtime": 9.3979,
      "eval_samples_per_second": 399.026,
      "eval_steps_per_second": 6.278,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03922512009739876,
      "learning_rate": 6.57197748161099e-05,
      "loss": 0.2884,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4439852833747864,
      "eval_runtime": 7.3675,
      "eval_samples_per_second": 508.989,
      "eval_steps_per_second": 8.008,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1432125568389893,
      "learning_rate": 3.481857044676613e-05,
      "loss": 0.1099,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.3750912845134735,
      "eval_runtime": 7.4193,
      "eval_samples_per_second": 505.439,
      "eval_steps_per_second": 7.952,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08836914598941803,
      "learning_rate": 9.71615616894559e-06,
      "loss": 0.0293,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9253333333333333,
      "eval_loss": 0.29097703099250793,
      "eval_runtime": 7.015,
      "eval_samples_per_second": 534.568,
      "eval_steps_per_second": 8.411,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03846749663352966,
      "learning_rate": 8.010370950384261e-10,
      "loss": 0.0046,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.2760196030139923,
      "eval_runtime": 7.092,
      "eval_samples_per_second": 528.769,
      "eval_steps_per_second": 8.319,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}