tokenizer_config.json

{
  "add_prefix_space": null,
  "backend": "tokenizers",
  "bos_token": "<｜begin▁of▁sentence｜>",
  "clean_up_tokenization_spaces": false,
  "eos_token": "<|EOT|>",
  "is_local": true,
  "model_max_length": 16384,
  "pad_token": "<｜end▁of▁sentence｜>",
  "sp_model_kwargs": {},
  "tokenizer_class": "LlamaTokenizer",
  "unk_token": null,
  "use_default_system_prompt": false
}