metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: token_type_ids
sequence: int8
- name: attention_mask
sequence: int8
- name: special_tokens_mask
sequence: int8
splits:
- name: train
num_bytes: 31188083
num_examples: 8622
- name: test
num_bytes: 3473125
num_examples: 958
- name: validation
num_bytes: 3473125
num_examples: 958
download_size: 1284779
dataset_size: 38134333
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*