metadata
dataset_info:
features:
- name: translation
struct:
- name: en
dtype: string
- name: mn
dtype: string
splits:
- name: train
num_bytes: 8697740
num_examples: 22350
- name: validation
num_bytes: 367443
num_examples: 1001
- name: test
num_bytes: 336049
num_examples: 1001
download_size: 4471836
dataset_size: 9401232
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*