---
dataset_info:
  features:
  - name: index
    dtype: int64
  - name: prompt
    dtype: string
  - name: system
    dtype: string
  - name: ground_truth
    dtype: string
  - name: scale
    dtype: string
  - name: post_id
    dtype: string
  - name: persona
    dtype: string
  - name: generation
    dtype: string
  - name: prediction
    dtype: string
  - name: accuracy
    dtype: float64
  - name: has_tags
    dtype: bool
  - name: scale_min
    dtype: int64
  - name: scale_max
    dtype: int64
  splits:
  - name: train
    num_bytes: 19042470
    num_examples: 10000
  download_size: 4004925
  dataset_size: 19042470
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---