trainer_state.json

{
  "best_metric": 0.9133333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0202/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": 29.37284278869629,
      "learning_rate": 4.003003003003003e-05,
      "loss": 0.9286,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.49114498496055603,
      "eval_runtime": 11.5635,
      "eval_samples_per_second": 324.296,
      "eval_steps_per_second": 5.102,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": NaN,
      "learning_rate": 3.0060060060060065e-05,
      "loss": 0.2456,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8805333333333333,
      "eval_loss": 0.38965585827827454,
      "eval_runtime": 11.6213,
      "eval_samples_per_second": 322.684,
      "eval_steps_per_second": 5.077,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 78.27859497070312,
      "learning_rate": 2.006006006006006e-05,
      "loss": 0.1012,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.37347251176834106,
      "eval_runtime": 11.581,
      "eval_samples_per_second": 323.807,
      "eval_steps_per_second": 5.095,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 8.925686836242676,
      "learning_rate": 1.006006006006006e-05,
      "loss": 0.0382,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.3536548912525177,
      "eval_runtime": 11.2368,
      "eval_samples_per_second": 333.726,
      "eval_steps_per_second": 5.251,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5969616770744324,
      "learning_rate": 6.006006006006006e-08,
      "loss": 0.01,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9133333333333333,
      "eval_loss": 0.3491658866405487,
      "eval_runtime": 11.3329,
      "eval_samples_per_second": 330.895,
      "eval_steps_per_second": 5.206,
      "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
}