dataset_info.json

{
  "split": "train",
  "total_samples": 7055,
  "total_shards": 7,
  "total_bytes": 6842224640,
  "missing_captions": 0
}