trainer_state.json

{
  "best_metric": 0.4304,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0196/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": 0.00027184967850358286,
      "loss": 3.2973,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22186666666666666,
      "eval_loss": 2.929622173309326,
      "eval_runtime": 5.9641,
      "eval_samples_per_second": 628.758,
      "eval_steps_per_second": 9.892,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00019742791503092223,
      "loss": 2.7093,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.296,
      "eval_loss": 2.644272565841675,
      "eval_runtime": 5.8158,
      "eval_samples_per_second": 644.799,
      "eval_steps_per_second": 10.145,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 35.58650207519531,
      "learning_rate": 0.00010472545707642648,
      "loss": 2.3552,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.36346666666666666,
      "eval_loss": 2.331881523132324,
      "eval_runtime": 5.9342,
      "eval_samples_per_second": 631.932,
      "eval_steps_per_second": 9.942,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 37.33841323852539,
      "learning_rate": 2.931633569361981e-05,
      "loss": 1.99,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4088,
      "eval_loss": 2.136383295059204,
      "eval_runtime": 5.9095,
      "eval_samples_per_second": 634.566,
      "eval_steps_per_second": 9.984,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 37.46784591674805,
      "learning_rate": 4.272188967840984e-09,
      "loss": 1.6782,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4304,
      "eval_loss": 2.090273141860962,
      "eval_runtime": 5.8651,
      "eval_samples_per_second": 639.375,
      "eval_steps_per_second": 10.06,
      "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
}