---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: chosen
    dtype: string
  - name: rejected
    dtype: string
  - name: source_dataset
    dtype: string
  splits:
  - name: train
    num_bytes: 870035129
    num_examples: 251356
  download_size: 503351311
  dataset_size: 870035129
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---