---
dataset_info:
  features:
  - name: data
    struct:
    - name: best_rank
      dtype: int32
    - name: best_response
      dtype: string
    - name: instruction
      dtype: string
    - name: worst_rank
      dtype: int32
    - name: worst_response
      dtype: string
  splits:
  - name: train
    num_bytes: 118097
    num_examples: 50
  download_size: 86893
  dataset_size: 118097
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---