---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: content
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: teacher_response
    dtype: string
  - name: category
    dtype: string
  - name: grounded
    dtype: bool
  - name: flaw
    dtype: string
  - name: agreement
    dtype: bool
  - name: combined_content
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  splits:
  - name: train
    num_bytes: 69481301
    num_examples: 31831
  download_size: 34033461
  dataset_size: 69481301
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---