metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 739721593504
num_examples: 364868892
- name: validation
num_bytes: 60051536
num_examples: 30000
download_size: 376936706386
dataset_size: 739781645040
