metadata
dataset_info:
features:
- name: protocol_id
dtype: string
- name: section_number
dtype: string
- name: title
dtype: string
- name: content
dtype: string
- name: merged_titles
list: string
splits:
- name: train
num_bytes: 47523272
num_examples: 18103
download_size: 20710041
dataset_size: 47523272
configs:
- config_name: default
data_files:
- split: train
path: data/train-*