metadata
dataset_info:
features:
- name: completion
dtype: string
- name: question_id
dtype: string
- name: prompt
dtype: string
- name: aligned
dtype: float64
- name: coherent
dtype: float64
splits:
- name: train
num_bytes: 34899.35152487962
num_examples: 53
download_size: 11210
dataset_size: 34899.35152487962
configs:
- config_name: default
data_files:
- split: train
path: data/train-*