README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: conversations
      dtype: string
  splits:
    - name: train
      num_bytes: 32172470547.488
      num_examples: 558128
  download_size: 27759109881
  dataset_size: 32172470547.488

Dataset Card for "llava-pretrain"

More Information needed