metadata
dataset_info:
features:
- name: document_label
dtype: int64
- name: sentence_labels
sequence: int64
- name: question_ids
sequence: int64
- name: response_ids
sequence: int64
- name: document_sentences_ids
sequence:
sequence: int64
- name: length
dtype: int64
splits:
- name: train
num_bytes: 2190654260
num_examples: 1349810
- name: test
num_bytes: 136684576
num_examples: 41504
- name: validation
num_bytes: 234049744
num_examples: 140286
download_size: 701502735
dataset_size: 2561388580
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*