metadata
dataset_info:
features:
- name: hi
dtype: string
- name: en
dtype: string
- name: en_embedding
sequence:
sequence:
sequence: float32
splits:
- name: train
num_bytes: 31578005
num_examples: 100
download_size: 32159877
dataset_size: 31578005
configs:
- config_name: default
data_files:
- split: train
path: data/train-*