---
dataset_info:
  features:
  - name: rationale
    dtype: string
  - name: answer
    dtype: string
  - name: question
    dtype: string
  - name: question_type
    dtype: string
  - name: meta
    struct:
    - name: entity
      dtype: string
    - name: fact
      dtype: string
    - name: hops
      dtype: int64
    - name: multi_hop_path
      sequence: string
    - name: reasoning_type
      dtype: string
  splits:
  - name: train
    num_bytes: 275790
    num_examples: 520
  download_size: 83433
  dataset_size: 275790
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---