metadata
dataset_info:
config_name: cleaned
features:
- name: doc_id
dtype: string
- name: o_body
dtype: string
- name: doc_name
dtype: string
- name: doc_source
dtype: string
- name: doc_type
dtype: string
- name: doc_format
dtype: string
- name: l_body
dtype: string
- name: p_body
dtype: string
splits:
- name: pretrain
num_bytes: 220630697
num_examples: 60480
download_size: 102561061
dataset_size: 220630697
configs:
- config_name: cleaned
data_files:
- split: pretrain
path: cleaned/pretrain-*