---
dataset_info:
  features:
  - name: metadata
    dtype: string
  - name: problem
    dtype: string
  - name: answer
    dtype: string
  - name: source_config
    dtype: string
  - name: model_response
    dtype: string
  splits:
  - name: train
    num_bytes: 329042
    num_examples: 500
  download_size: 175276
  dataset_size: 329042
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---