tokenizer_config.json

{
  "added_tokens_decoder": {
    "31996": {
      "content": "<|startoftext|>",
      "lstrip": false,
      "normalized": false,
      "rstrip": false,
      "single_word": false,
      "special": true
    },
    "31999": {
      "content": "<|endoftext|>",
      "lstrip": false,
      "normalized": false,
      "rstrip": false,
      "single_word": false,
      "special": true
    }
  },
  "bos_token": "<|startoftext|>",
  "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '\n### User:\n' }}{{ message['content']}}{{ '\n### Assistant:\n' }}{% endif %}{% if message['role'] == 'assistant' %}{{ message['content']+eos_token}}{% endif %}{% endfor %}",
  "clean_up_tokenization_spaces": true,
  "do_clean_text": false,
  "eos_token": "<|endoftext|>",
  "model_max_length": 2048,
  "pad_token": "<|endoftext|>",
  "tokenizer_class": "GPTNeoXJapaneseTokenizer",
  "unk_token": "<|endoftext|>"
}