metadata
language:
- en
dataset_info:
features:
- name: query
dtype: string
- name: document
dtype: string
- name: dataset
dtype: string
- name: shard
dtype: int64
splits:
- name: train
num_bytes: 96427771748.06602
num_examples: 47799699
download_size: 52219022280
dataset_size: 96427771748.06602
configs:
- config_name: default
data_files:
- split: train
path: data/train-*