---
dataset_info:
  features:
  - name: instance_id
    dtype: string
  - name: context
    dtype: string
  - name: correct_response
    dtype: string
  - name: response_a
    dtype: string
  - name: response_b
    dtype: string
  - name: difficulty
    dtype: float64
  splits:
  - name: train
    num_bytes: 60368619
    num_examples: 9574
  download_size: 25469876
  dataset_size: 60368619
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---