README.md

metadata

dataset_info:
  features:
    - name: text
      dtype: string
    - name: subset
      dtype: string
  splits:
    - name: train
      num_bytes: 22784812902
      num_examples: 5000000
  download_size: 13920512648
  dataset_size: 22784812902
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*