metadata
dataset_info:
features:
- name: path
dtype: string
- name: sentence
dtype: string
- name: array
sequence: float32
splits:
- name: train
num_bytes: 1892874708
num_examples: 7183
- name: test
num_bytes: 1466625438
num_examples: 5015
download_size: 3345740901
dataset_size: 3359500146
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*