README.md

metadata

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-*