metadata
dataset_info:
features:
- name: en
dtype: string
- name: vi
dtype: string
- name: source
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 2423142843
num_examples: 2884451
- name: validation
num_bytes: 9990791
num_examples: 11316
- name: test
num_bytes: 9925956
num_examples: 11225
download_size: 810147374
dataset_size: 2443059590
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*