metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: label
dtype: bool
- name: bertscore_f1
dtype: float64
- name: rank
dtype: int64
- name: file_name
dtype: string
- name: categories
dtype: string
- name: subcategories
dtype: string
- name: reference_explanation
dtype: string
splits:
- name: train
num_bytes: 1770870
num_examples: 440
download_size: 266488
dataset_size: 1770870
configs:
- config_name: default
data_files:
- split: train
path: data/train-*