---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: quality
    list:
    - name: status
      dtype: string
    - name: user_id
      dtype: string
    - name: value
      dtype: string
  - name: metadata
    dtype: string
  - name: avg_rating
    dtype: float64
  - name: num_responses
    dtype: int64
  - name: agreement_ratio
    dtype: float64
  - name: raw_responses
    sequence: int64
  - name: kind
    dtype: string
  splits:
  - name: train
    num_bytes: 10601573
    num_examples: 10331
  download_size: 4323988
  dataset_size: 10601573
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---