---
dataset_info:
  features:
  - name: prompts
    dtype: string
  - name: responses_1
    dtype: string
  - name: responses_2
    dtype: string
  - name: responses_merged
    dtype: string
  splits:
  - name: train
    num_bytes: 1631670
    num_examples: 4000
  download_size: 499953
  dataset_size: 1631670
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---