---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: text
    dtype: string
  - name: embeddings
    list: float32
  splits:
  - name: train
    num_bytes: 15837286488.275
    num_examples: 291785
  download_size: 48314519876
  dataset_size: 15837286488.275
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---