metadata
dataset_info:
features:
- name: Sentences
dtype: string
- name: Questions
dtype: string
- name: __index_level_0__
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 525733
num_examples: 663
download_size: 16387
dataset_size: 525733
configs:
- config_name: default
data_files:
- split: train
path: data/train-*