metadata
dataset_info:
features:
- name: source_lang
dtype: string
- name: target_lang
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
- name: data_split
dtype: string
- name: data_source
dtype: string
splits:
- name: train
num_bytes: 42764837609.76546
num_examples: 193390351
- name: validation
num_bytes: 431968081.23454416
num_examples: 1953438
download_size: 23665956443
dataset_size: 43196805691
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*