---
dataset_info:
  features:
  - name: prompt
    dtype: string
  - name: meta
    struct:
    - name: generation
      dtype: string
    - name: generation_method
      dtype: string
    - name: group
      dtype: string
    - name: prompt
      dtype: string
    - name: prompt_label
      dtype: int64
    - name: roberta_prediction
      dtype: float64
  - name: response
    dtype: string
  - name: attribute_name
    dtype: string
  - name: attribute_value
    dtype: int64
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 340656458
    num_examples: 250951
  download_size: 36007442
  dataset_size: 340656458
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---