metadata
dataset_info:
features:
- name: title
dtype: string
- name: text
dtype: string
- name: bm25_text
dtype: string
splits:
- name: train
num_bytes: 2889457164
num_examples: 1944406
download_size: 1242752879
dataset_size: 2889457164
configs:
- config_name: default
data_files:
- split: train
path: data/train-*