README.md

---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: input
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: output
    struct:
    - name: thinking
      dtype: string
    - name: answer
      dtype: string
  - name: reasoning
    dtype: string
  - name: generator
    dtype: string
  - name: difficulty_level
    dtype: int64
  - name: required_search_score
    dtype: float64
  - name: category
    dtype: string
  - name: source_urls
    list: 'null'
  - name: license
    dtype: string
  - name: question_source
    dtype: string
  - name: dataset_id
    dtype: string
  - name: language
    dtype: string
  splits:
  - name: train
    num_bytes: 247690114
    num_examples: 31043
  download_size: 106441676
  dataset_size: 247690114
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---