metadata
dataset_info:
features:
- name: seq
dtype: string
- name: median_count
dtype: float32
splits:
- name: train
num_bytes: 21039298344
num_examples: 20386917
- name: validation
num_bytes: 1168849392
num_examples: 1132606
- name: test
num_bytes: 1168851456
num_examples: 1132608
download_size: 11643417229
dataset_size: 23376999192
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*