tokenizer_config.json

{
  "tokenizer_class": "BPETokenizer",
  "vocab_size": 4000,
  "model_type": "bpe",
  "special_tokens": {
    "<PAD>": 0,
    "<UNK>": 1,
    "<BOS>": 2,
    "<EOS>": 3
  },
  "do_lower_case": true,
  "language": "id"
}