---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: caption
    dtype: string
  - name: title
    dtype: string
  - name: vlm_caption
    dtype: string
  - name: original_text
    dtype: string
  - name: source
    dtype: string
  - name: board
    dtype: string
  - name: doc_id
    dtype: string
  - name: url
    dtype: string
  - name: timestamp
    dtype: string
  splits:
  - name: train
    num_bytes: 81056332
    num_examples: 200
  download_size: 80172780
  dataset_size: 81056332
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---