---
dataset_info:
  features:
  - name: score
    dtype: float32
  - name: Kongo
    dtype: string
  - name: Lingala
    dtype: string
  - name: Swahili
    dtype: string
  - name: similarity
    dtype: float32
  - name: French
    dtype: string
  - name: English
    dtype: string
  splits:
  - name: train
    num_bytes: 858073679
    num_examples: 7043744
  download_size: 518255401
  dataset_size: 858073679
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---