metadata
dataset_info:
features:
- name: document
dtype: string
- name: summary
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 670495326
num_examples: 20000
- name: validation
num_bytes: 33456233
num_examples: 1000
- name: test
num_bytes: 33942448
num_examples: 1000
download_size: 376078489
dataset_size: 737894007
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*