tokenizer/tokenizer_config.json

{
  "add_prefix_space": false,
  "added_tokens_decoder": {
    "50256": {
      "content": "<|endoftext|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": true
    },
    "50257": {
      "content": "<|start_of_turn|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": false
    },
    "50258": {
      "content": "<|end_of_turn|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": false
    },
    "50259": {
      "content": "<|user|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": false
    },
    "50260": {
      "content": "<|assistant|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": false
    },
    "50261": {
      "content": "<|system|>",
      "lstrip": false,
      "normalized": true,
      "rstrip": false,
      "single_word": false,
      "special": false
    }
  },
  "bos_token": "<|endoftext|>",
  "chat_template": "{{ bos_token }}\n{% for message in messages %}\n    {% if (message['role'] != 'assistant' and message['role'] != 'user' and message['role'] != 'system') %}\n        {{ raise_exception('Only user and assistant and system role are supported') }}\n    {% endif %}\n    {{ '<|start_of_turn|>' + '<|' + message['role'] + '|>'+ '\\n' + message['content'] | trim + '<|end_of_turn|>\\n' }}\n{% endfor %}\n{% if add_generation_prompt %}\n    {{'<|start_of_turn|><|assistant|>\\n'}}\n{% endif %}",
  "clean_up_tokenization_spaces": false,
  "eos_token": "<|endoftext|>",
  "model_max_length": 1024,
  "pad_token": "<|endoftext|>",
  "tokenizer_class": "GPT2Tokenizer",
  "unk_token": "<|endoftext|>"
}