Datasets:

vonewman
/

french-instruction-dataset

dataset_info:
  features:
    - name: conversations
      list:
        - name: from
          dtype: string
        - name: value
          dtype: string
    - name: source
      dtype: string
  splits:
    - name: train
      num_bytes: 204585375
      num_examples: 108991
  download_size: 100056790
  dataset_size: 204585375
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: apache-2.0
language:
  - fr