metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 1848605112
num_examples: 1798254
- name: validation
num_bytes: 16842752
num_examples: 16384
download_size: 519090631
dataset_size: 1865447864
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*