---
dataset_info:
  features:
  - name: images
    sequence: image
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 152248793455.6
    num_examples: 950
  - name: test
    num_bytes: 8013094392.4
    num_examples: 50
  download_size: 164605519
  dataset_size: 160261887848.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---