metadata
dataset_info:
features:
- name: added
dtype: string
- name: created
dtype: timestamp[s]
- name: id
dtype: string
- name: metadata
struct:
- name: authors
dtype: string
- name: license
dtype: string
- name: provenance
dtype: string
- name: submitter
dtype: string
- name: url
dtype: string
- name: source
dtype: string
- name: text
dtype: string
- name: category
dtype: string
- name: token_length
dtype: int64
splits:
- name: train
num_bytes: 19589633023
num_examples: 295424
download_size: 9532573556
dataset_size: 19589633023
configs:
- config_name: default
data_files:
- split: train
path: data/train-*