README.md

---
tags:
- not-for-all-audiences
dataset_info:
  features:
  - name: prompt_id
    dtype: string
  - 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
    num_bytes: 128690951
    num_examples: 85074
  - name: test
    num_bytes: 7201288
    num_examples: 4624
  download_size: 44628148
  dataset_size: 135892239
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
license: apache-2.0
task_categories:
- conversational
language:
- en
size_categories:
- 10K<n<100K
---