---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: answer
    dtype: string
  - name: mask
    dtype: bool
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 110272782
    num_examples: 25041
  download_size: 56276221
  dataset_size: 110272782
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---