metadata
dataset_info:
features:
- name: text
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 865791958
num_examples: 150523
- name: validation
num_bytes: 111633790
num_examples: 18815
download_size: 157402763
dataset_size: 977425748
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*