metadata
dataset_info:
features:
- name: exemplar_questions
dtype: string
- name: test_questions
dtype: string
- name: subject
dtype: string
- name: answers
list:
- name: answer
dtype: string
- name: model
dtype: string
- name: reference_answers
dtype: int64
splits:
- name: train
num_bytes: 279310672
num_examples: 1000
download_size: 10943728
dataset_size: 279310672
configs:
- config_name: default
data_files:
- split: train
path: data/train-*