metadata
dataset_info:
features:
- name: query
dtype: string
- name: document
dtype: string
- name: dataset
dtype: string
- name: shard
dtype: int64
splits:
- name: train
num_bytes: 482138856723
num_examples: 238998494
download_size: 260607608502
dataset_size: 482138856723
configs:
- config_name: default
data_files:
- split: train
path: data/train-*