---
dataset_info:
  features:
  - name: chunks
    struct:
    - name: retrieved
      sequence: string
  - name: ids
    dtype: string
  splits:
  - name: train
    num_bytes: 13778479891
    num_examples: 3236695
  download_size: 7469710102
  dataset_size: 13778479891
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---