metadata
dataset_info:
features:
- name: system
dtype: string
- name: prompt
dtype: string
- name: answer
dtype: string
- name: correct
dtype: int64
splits:
- name: train
num_bytes: 2813612.7
num_examples: 82
download_size: 24939
dataset_size: 2813612.7
configs:
- config_name: default
data_files:
- split: train
path: data/train-*