---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: subset
    dtype: string
  splits:
  - name: train
    num_bytes: 22784812902.0
    num_examples: 5000000
  download_size: 13920512648
  dataset_size: 22784812902.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---