---
dataset_info:
  features:
  - name: chunks
    dtype: string
  splits:
  - name: train
    num_bytes: 19406680850
    num_examples: 22300752
  download_size: 11345406685
  dataset_size: 19406680850
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---