---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: input_image
    dtype: image
  - name: weak_prompt
    dtype: string
  - name: verdict
    dtype:
      class_label:
        names:
          '0': 'False'
          '1': 'True'
  - name: ground_truth_reasoning
    dtype: string
  splits:
  - name: train
    num_bytes: 206966837
    num_examples: 33
  download_size: 206959328
  dataset_size: 206966837
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---