---
dataset_info:
  features:
  - name: english
    dtype: string
  - name: eng_token_len
    dtype: int64
  - name: korean
    dtype: string
  - name: ko_token_len
    dtype: int64
  - name: token_len_diff
    dtype: int64
  splits:
  - name: train
    num_bytes: 5134954835
    num_examples: 500000
  download_size: 2664684446
  dataset_size: 5134954835
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---