metadata
dataset_info:
features:
- name: id
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: answers
list: string
- name: dataset_name
dtype: string
- name: domain
dtype: string
- name: num_context_tokens
dtype: int64
splits:
- name: train
num_bytes: 161803442
num_examples: 32479
download_size: 46344508
dataset_size: 161803442
configs:
- config_name: default
data_files:
- split: train
path: data/train-*