---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: caption
    dtype: string
  - name: image_name
    dtype: string
  - name: source_folder
    dtype: string
  - name: responses
    dtype: string
  splits:
  - name: train
    num_bytes: 510122513.625
    num_examples: 1491
  download_size: 507455883
  dataset_size: 510122513.625
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---