metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 1295254968
num_examples: 126391
- name: validation
num_bytes: 111959400
num_examples: 10925
- name: test
num_bytes: 109510128
num_examples: 10686
download_size: 288396493
dataset_size: 1516724496
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*