metadata
dataset_info:
features:
- name: summary
dtype: string
- name: text
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 11563216633
num_examples: 32747
- name: test
num_bytes: 3549727516
num_examples: 10557
- name: validation
num_bytes: 1213219924
num_examples: 3461
download_size: 6300546543
dataset_size: 16326164073
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*