Datasets:

Svngoku
/

Inkuba-Swahili-MMT

dataset_info:
  features:
    - name: instruction
      dtype: string
    - name: inputs
      dtype: string
    - name: targets
      dtype: string
    - name: task
      dtype: string
    - name: data_source
      dtype: string
  splits:
    - name: train
      num_bytes: 41514508.256710336
      num_examples: 200000
  download_size: 19791456
  dataset_size: 41514508.256710336
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
language:
  - sw
  - en
tags:
  - translation
size_categories:
  - 100K<n<1M