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: 24061983205.32434
num_examples: 11927638
download_size: 13340024853
dataset_size: 24061983205.32434
configs:
- config_name: default
data_files:
- split: train
path: data/train-*