---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: translation
    dtype:
      translation:
        languages:
        - vi
        - zh
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 5673312936
    num_examples: 13329734
  - name: test
    num_bytes: 630033778
    num_examples: 1481082
  download_size: 3190585179
  dataset_size: 6303346714
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---