metadata
dataset_info:
features:
- name: sequence
dtype: string
splits:
- name: train
num_bytes: 26063849199.91452
num_examples: 84987726
- name: valid
num_bytes: 3066778.0427393154
num_examples: 10000
- name: test
num_bytes: 3066778.0427393154
num_examples: 10000
download_size: 26147724298
dataset_size: 26069982756
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*