metadata
dataset_info:
features:
- name: added
dtype: string
- name: created
dtype: timestamp[us]
- name: id
dtype: string
- name: metadata
struct:
- name: all_licenses
sequence: string
- name: authors
sequence: string
- name: include_comments
dtype: bool
- name: license
dtype: string
- name: provenance
dtype: string
- name: site
dtype: string
- name: sort
dtype: string
- name: url
dtype: string
- name: source
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 17778208427
num_examples: 5000000
download_size: 8732125312
dataset_size: 17778208427
configs:
- config_name: default
data_files:
- split: train
path: data/train-*