---
dataset_info:
  features:
  - name: source_id
    dtype: int64
  - name: question
    dtype: string
  - name: response
    dtype: string
  - name: metadata
    struct:
    - name: source
      sequence: string
  splits:
  - name: train
    num_bytes: 125049216
    num_examples: 46860
  download_size: 64816414
  dataset_size: 125049216
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---