---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: instruction
    dtype: string
  - name: qtype
    dtype: string
  - name: pubid
    dtype: int32
  splits:
  - name: train
    num_bytes: 9543193566
    num_examples: 21608801
  download_size: 5032171833
  dataset_size: 9543193566
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---