---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: corpus
    dtype: string
  - name: original_id
    dtype: int64
  splits:
  - name: train
    num_bytes: 141807806497
    num_examples: 50336214
  download_size: 84893303434
  dataset_size: 141807806497
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---