---
dataset_info:
  features:
  - name: conversation_id
    dtype: string
  - name: message_id
    dtype: string
  - name: user_prompt
    dtype: string
  - name: assistant_response
    dtype: string
  - name: images
    sequence: image
  - name: num_images
    dtype: int64
  splits:
  - name: train
    num_bytes: 167951747.0
    num_examples: 73
  download_size: 167808065
  dataset_size: 167951747.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---