metadata
dataset_info:
features:
- name: context
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: answer_prefix
dtype: string
- name: max_new_tokens
dtype: int64
- name: tuple_count
dtype: int64
- name: token_count
dtype: int64
splits:
- name: test
num_bytes: 3648797
num_examples: 1078
download_size: 678706
dataset_size: 3648797
configs:
- config_name: default
data_files:
- split: test
path: data/test-*