README.md

---
dataset_info:
  features:
  - name: prompt
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: data_source
    dtype: string
  - name: ability
    dtype: string
  - name: answer
    dtype: string
  - name: raw_problem
    dtype: string
  - name: level
    dtype: int64
  - name: type
    dtype: string
  - name: reward_model
    struct:
    - name: ground_truth
      dtype: string
    - name: style
      dtype: string
  - name: extra_info
    struct:
    - name: problem_id
      dtype: string
  splits:
  - name: train
    num_bytes: 9518
    num_examples: 1
  download_size: 10503
  dataset_size: 9518
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---