---
dataset_info:
  features:
  - name: images
    list: image
  - name: texts
    list:
    - name: user
      dtype: string
    - name: assistant
      dtype: string
    - name: source
      dtype: string
  splits:
  - name: train
    num_bytes: 132298787202
    num_examples: 155865
  download_size: 131689986076
  dataset_size: 132298787202
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---