metadata
dataset_info:
features:
- name: docid
dtype: string
- name: title
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 483600907.4401386
num_examples: 1301683
- name: test
num_bytes: 53733640.559861444
num_examples: 144632
download_size: 345590743
dataset_size: 537334548
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*