metadata
dataset_info:
features:
- name: text
dtype: string
- name: pdf
dtype: binary
splits:
- name: train
num_bytes: 8425352699
num_examples: 87295
download_size: 7641642477
dataset_size: 8425352699
configs:
- config_name: default
data_files:
- split: train
path: data/train-*