---
dataset_info:
  features:
  - name: doc_id
    dtype: string
  - name: sent_idx
    dtype: int64
  - name: text
    dtype: string
  - name: source
    dtype: string
  - name: domain
    dtype: string
  - name: detected_lang
    dtype: string
  - name: lang_confidence
    dtype: float64
  - name: num_chars
    dtype: int64
  - name: is_kk
    dtype: bool
  splits:
  - name: train
    num_bytes: 47197488
    num_examples: 135808
  download_size: 20618813
  dataset_size: 47197488
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---