metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 1991070700
num_examples: 485627
download_size: 287527718
dataset_size: 1991070700
configs:
- config_name: default
data_files:
- split: train
path: data/train-*