metadata
license: mit
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 618759636
num_examples: 47553
- name: test
num_bytes: 207151040
num_examples: 15920
- name: validation
num_bytes: 450735680
num_examples: 34640
download_size: 279866921
dataset_size: 1276646356