---
dataset_info:
  features:
  - name: prompt
    dtype: large_string
  - name: completion
    dtype: large_string
  splits:
  - name: train
    num_bytes: 20059391
    num_examples: 5545
  - name: variation
    num_bytes: 8708
    num_examples: 500
  download_size: 2410336
  dataset_size: 20068099
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: variation
    path: data/variation-*
---