metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 51420
num_examples: 800
- name: eval
num_bytes: 12860
num_examples: 200
download_size: 7087
dataset_size: 64280
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*