README.md

metadata

dataset_info:
  features:
    - name: ar
      dtype: string
    - name: en
      dtype: string
    - name: sco
      dtype: float64
  splits:
    - name: train
      num_bytes: 37475572129
      num_examples: 115355613
  download_size: 22348241927
  dataset_size: 37475572129
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*