metadata
dataset_info:
config_name: ar
features:
- name: prompt
dtype: string
- name: query_id
dtype: string
- name: doc_ids
sequence: string
- name: documents
list:
- name: docid
dtype: string
- name: text
dtype: string
- name: title
dtype: string
- name: gold_answer
sequence: string
splits:
- name: train
num_bytes: 102496608
num_examples: 6500
download_size: 50730638
dataset_size: 102496608
configs:
- config_name: ar
data_files:
- split: train
path: ar/train-*