Narrative-Brain-5B-NF4 / tokenizer_config.json
Veetance's picture
Add Narrative Brain tokenizer config
a93fea3 verified
Raw History Blame Contribute Delete
20.3 kB
{
"add_bos_token": false,
"add_prefix_space": false,
"added_tokens_decoder": {
"151643": {
"content": "\u003c|endoftext|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151644": {
"content": "\u003c|im_start|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151645": {
"content": "\u003c|im_end|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151646": {
"content": "\u003c|object_ref_start|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151647": {
"content": "\u003c|object_ref_end|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151648": {
"content": "\u003c|box_start|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151649": {
"content": "\u003c|box_end|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151650": {
"content": "\u003c|quad_start|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151651": {
"content": "\u003c|quad_end|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151652": {
"content": "\u003c|vision_start|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151653": {
"content": "\u003c|vision_end|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151654": {
"content": "\u003c|vision_pad|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151655": {
"content": "\u003c|image_pad|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151656": {
"content": "\u003c|video_pad|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"151657": {
"content": "\u003ctool_call\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151658": {
"content": "\u003c/tool_call\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151659": {
"content": "\u003c|fim_prefix|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151660": {
"content": "\u003c|fim_middle|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151661": {
"content": "\u003c|fim_suffix|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151662": {
"content": "\u003c|fim_pad|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151663": {
"content": "\u003c|repo_name|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151664": {
"content": "\u003c|file_sep|\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151665": {
"content": "\u003ctool_response\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151666": {
"content": "\u003c/tool_response\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151667": {
"content": "\u003cthink\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
},
"151668": {
"content": "\u003c/think\u003e",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": false
}
},
"additional_special_tokens": [
"\u003c|im_start|\u003e",
"\u003c|im_end|\u003e",
"\u003c|object_ref_start|\u003e",
"\u003c|object_ref_end|\u003e",
"\u003c|box_start|\u003e",
"\u003c|box_end|\u003e",
"\u003c|quad_start|\u003e",
"\u003c|quad_end|\u003e",
"\u003c|vision_start|\u003e",
"\u003c|vision_end|\u003e",
"\u003c|vision_pad|\u003e",
"\u003c|image_pad|\u003e",
"\u003c|video_pad|\u003e"
],
"bos_token": null,
"chat_template": "{%- if tools %}\n {{- \u0027\u003c|im_start|\u003esystem\\n\u0027 }}\n {%- if messages[0].role == \u0027system\u0027 %}\n {{- messages[0].content + \u0027\\n\\n\u0027 }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within \u003ctools\u003e\u003c/tools\u003e XML tags:\\n\u003ctools\u003e\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n\u003c/tools\u003e\\n\\nFor each function call, return a json object with function name and arguments within \u003ctool_call\u003e\u003c/tool_call\u003e XML tags:\\n\u003ctool_call\u003e\\n{\\\"name\\\": \u003cfunction-name\u003e, \\\"arguments\\\": \u003cargs-json-object\u003e}\\n\u003c/tool_call\u003e\u003c|im_end|\u003e\\n\" }}\n{%- else %}\n {%- if messages[0].role == \u0027system\u0027 %}\n {{- \u0027\u003c|im_start|\u003esystem\\n\u0027 + messages[0].content + \u0027\u003c|im_end|\u003e\\n\u0027 }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith(\u0027\u003ctool_response\u003e\u0027) and message.content.endswith(\u0027\u003c/tool_response\u003e\u0027)) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = \u0027\u0027 %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- \u0027\u003c|im_start|\u003e\u0027 + message.role + \u0027\\n\u0027 + content + \u0027\u003c|im_end|\u003e\u0027 + \u0027\\n\u0027 }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = \u0027\u0027 %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if \u0027\u003c/think\u003e\u0027 in content %}\n {%- set reasoning_content = content.split(\u0027\u003c/think\u003e\u0027)[0].rstrip(\u0027\\n\u0027).split(\u0027\u003cthink\u003e\u0027)[-1].lstrip(\u0027\\n\u0027) %}\n {%- set content = content.split(\u0027\u003c/think\u003e\u0027)[-1].lstrip(\u0027\\n\u0027) %}\n {%- endif %}\n {%- endif %}\n {%- if loop.index0 \u003e ns.last_query_index %}\n {%- if loop.last or (not loop.last and reasoning_content) %}\n {{- \u0027\u003c|im_start|\u003e\u0027 + message.role + \u0027\\n\u003cthink\u003e\\n\u0027 + reasoning_content.strip(\u0027\\n\u0027) + \u0027\\n\u003c/think\u003e\\n\\n\u0027 + content.lstrip(\u0027\\n\u0027) }}\n {%- else %}\n {{- \u0027\u003c|im_start|\u003e\u0027 + message.role + \u0027\\n\u0027 + content }}\n {%- endif %}\n {%- else %}\n {{- \u0027\u003c|im_start|\u003e\u0027 + message.role + \u0027\\n\u0027 + content }}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- \u0027\\n\u0027 }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- \u0027\u003ctool_call\u003e\\n{\"name\": \"\u0027 }}\n {{- tool_call.name }}\n {{- \u0027\", \"arguments\": \u0027 }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- \u0027}\\n\u003c/tool_call\u003e\u0027 }}\n {%- endfor %}\n {%- endif %}\n {{- \u0027\u003c|im_end|\u003e\\n\u0027 }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- \u0027\u003c|im_start|\u003euser\u0027 }}\n {%- endif %}\n {{- \u0027\\n\u003ctool_response\u003e\\n\u0027 }}\n {{- content }}\n {{- \u0027\\n\u003c/tool_response\u003e\u0027 }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- \u0027\u003c|im_end|\u003e\\n\u0027 }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- \u0027\u003c|im_start|\u003eassistant\\n\u0027 }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- \u0027\u003cthink\u003e\\n\\n\u003c/think\u003e\\n\\n\u0027 }}\n {%- endif %}\n{%- endif %}",
"clean_up_tokenization_spaces": false,
"eos_token": "\u003c|im_end|\u003e",
"errors": "replace",
"model_max_length": 32768,
"pad_token": "\u003c|endoftext|\u003e",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null,
"is_local": true
}