metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: eval
path: data/eval-*
dataset_info:
features:
- name: en
dtype: string
- name: vi
dtype: string
- name: loss
dtype: float64
splits:
- name: train
num_bytes: 1343163221
num_examples: 4163853
- name: test
num_bytes: 336407
num_examples: 1268
- name: eval
num_bytes: 353494
num_examples: 1553
download_size: 208876651
dataset_size: 1343853122