---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: completion
    dtype: string
  - name: clean_completion
    dtype: string
  - name: thoughts
    dtype: string
  - name: answer
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 17347426
    num_examples: 760
  download_size: 3358388
  dataset_size: 17347426
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---