metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: int64
- name: model_answer
dtype: string
- name: background_info
dtype: string
splits:
- name: train
num_bytes: 337066
num_examples: 1000
download_size: 7796
dataset_size: 337066
configs:
- config_name: default
data_files:
- split: train
path: data/train-*