README.md

---
dataset_info:
  features:
  - name: path
    dtype: string
  - name: question
    dtype: string
  - name: response
    dtype: string
  - name: extracted_answer
    dtype: string
  - name: model
    dtype: string
  - name: valid
    dtype: bool
  - name: list_failed
    sequence: 'null'
  - name: is_valid_end
    dtype: bool
  - name: 'Unnamed: 0'
    dtype: int64
  - name: Prompt
    dtype: string
  - name: Answer
    dtype: string
  - name: wikipedia_link_11+
    dtype: string
  - name: reasoning_types
    dtype: string
  - name: wiki_links
    dtype: string
  - name: answer_think
    dtype: string
  - name: is_correct
    dtype: string
  - name: refine_answer
    dtype: string
  - name: is_correct_pro_judge
    struct:
    - name: Decision
      dtype: string
    - name: Explanation
      dtype: string
  - name: is_correct_pro_judge_with_raw
    struct:
    - name: Decision
      dtype: string
    - name: Explanation
      dtype: string
  - name: refine_answer_pro
    dtype: string
  - name: is_correct_combine
    dtype: string
  - name: ground-truth
    dtype: string
  splits:
  - name: train
    num_bytes: 127602875
    num_examples: 823
  download_size: 59839437
  dataset_size: 127602875
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---