---
dataset_info:
  features:
  - name: images
    list: image
  - name: conversations
    list:
    - name: user
      dtype: string
    - name: assistant
      dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 2012107
    num_examples: 50
  download_size: 1931454
  dataset_size: 2012107
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---