update_chat_template_and_tokenizer

chat_template.jinja

@@ -1,12 +1,10 @@
 {%- if tools -%}
-  <|im_system|>tool_declare<|im_middle|>
-  # Tools
-  {{ tools | tojson }}<|im_end|>
+  <|im_system|>tool_declare<|im_middle|>{{ tools | tojson(separators=(',', ':')) }}<|im_end|>
 {%- endif -%}
-{%- for message in messages -%}
+{% for message in messages %}
   {%- if loop.first and messages[0]['role'] != 'system' -%}
-    <|im_system|>system<|im_middle|>You are Kimi, an AI assistant created by Moonshot AI.<|im_end|>
-  {%- endif -%}
+  <|im_system|>system<|im_middle|>You are Kimi, an AI assistant created by Moonshot AI.<|im_end|>
+  {% endif %}
   
   {%- set role_name =  message.get('name') or  message['role'] -%}
   {%- if message['role'] == 'user' -%}
@@ -15,7 +13,7 @@
     <|im_assistant|>{{role_name}}<|im_middle|>
   {%- else -%}
     <|im_system|>{{role_name}}<|im_middle|>
-  {% endif %}
+  {%- endif -%}
 
   {%- if message['role'] == 'assistant' and message.get('tool_calls') -%}
     {%- if message['content'] -%}{{ message['content'] }}{%- endif -%}
@@ -27,7 +25,7 @@
     <|tool_calls_section_end|>
   {%- elif message['role'] == 'tool' -%}
     ## Return of {{ message.tool_call_id }}
-    {{ message['content'] }}
+{{ message['content'] }}
   {%- elif message['content'] is string -%}
     {{ message['content'] }}
   {%- elif message['content'] is not none -%}

tokenization_kimi.py

@@ -17,12 +17,13 @@ from tiktoken.load import load_tiktoken_bpe
 from tokenizers import AddedToken, pre_tokenizers, Regex
 from transformers.tokenization_utils import PreTrainedTokenizer
 from transformers.models.gpt2.tokenization_gpt2 import bytes_to_unicode
-
+from typing import Any
 
 
 logger = getLogger(__name__)
 VOCAB_FILES_NAMES = {"vocab_file": "tiktoken.model"}
 
+
 class TikTokenTokenizer(PreTrainedTokenizer):
     """
     Tokenizing and encoding/decoding text using the Tiktoken tokenizer. See megatron/tokenizer/tiktoken_tokenizer.py.
@@ -321,3 +322,20 @@ class TikTokenTokenizer(PreTrainedTokenizer):
             copyfile(self.vocab_file, out_vocab_file)
 
         return (out_vocab_file,)
+    
+
+    
+    def apply_chat_template(
+        self, conversation, tools: Optional[list[dict]] = None, **kwargs
+    ):
+        tools = deep_sort_dict(tools)
+        return super().apply_chat_template(conversation, tools=tools, **kwargs)
+    
+
+def deep_sort_dict(obj: Any) -> Any:
+    if isinstance(obj, dict):
+        return {k: deep_sort_dict(v) for k, v in sorted(obj.items())}
+    if isinstance(obj, list):
+        return [deep_sort_dict(item) for item in obj]
+    return obj
+