---
dataset_info:
  features:
  - name: anchor
    dtype: string
  - name: chosen
    dtype: string
  - name: original_id
    dtype: string
  - name: category
    dtype: string
  - name: language
    dtype: string
  splits:
  - name: train
    num_bytes: 213108656
    num_examples: 669667
  download_size: 134996584
  dataset_size: 213108656
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---