Datasets:
metadata
dataset_info:
features:
- name: pageid
dtype: int64
- name: title
dtype: string
- name: revid
dtype: int64
- name: description
dtype: string
- name: categories
sequence: string
- name: inputs
dtype: string
- name: targets
dtype: string
splits:
- name: train
num_bytes: 829905155
num_examples: 44754
download_size: 489718761
dataset_size: 829905155
configs:
- config_name: default
data_files:
- split: train
path: data/train-*