metadata
dataset_info:
features:
- name: seq
dtype: string
- name: median_count
dtype: float32
splits:
- name: train
num_bytes: 20268889160
num_examples: 38978633
- name: validation
num_bytes: 1126049080
num_examples: 2165479
- name: test
num_bytes: 1126050120
num_examples: 2165481
download_size: 11105735970
dataset_size: 22520988360
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*