---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: chosen
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: rejected
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  splits:
  - name: train_split_1
    num_bytes: 12339488.511539096
    num_examples: 8484
  - name: train_split_2
    num_bytes: 12339488.511539096
    num_examples: 8484
  - name: train_split_3
    num_bytes: 12339488.511539096
    num_examples: 8484
  - name: train_split_4
    num_bytes: 12339488.511539096
    num_examples: 8484
  download_size: 27694116
  dataset_size: 49357954.046156384
configs:
- config_name: default
  data_files:
  - split: train_split_1
    path: data/train_split_1-*
  - split: train_split_2
    path: data/train_split_2-*
  - split: train_split_3
    path: data/train_split_3-*
  - split: train_split_4
    path: data/train_split_4-*
---