groupuser/Llama-3.2-1B-Instruct
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update 1/model.py

Browse Source
This commit is contained in:
cheetahadmin 2025-11-24 00:48:49 +00:00
parent 94cc13f1ea
commit dd748733bd
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
1/model.py
View File

@ -98,6 +98,10 @@ class TritonPythonModel:
self.tokenizer.chat_template is not None
)
self.logger.log_info(f"Supports Chat Template: {self.supports_chat_template}")
if self.supports_chat_template:
self.logger.log_info(f"Chat Template Content:\n{self.tokenizer.chat_template}")
def _get_bnb_config(self) -> Union[BitsAndBytesConfig, None]:
if self.quantization == "int4":
return BitsAndBytesConfig(
@ -131,7 +135,7 @@ class TritonPythonModel:
# [LOGGING] Request ID 포함하여 로그 출력
log_input_str = json.dumps(input_data, ensure_ascii=False) if isinstance(input_data, (list, dict)) else str(input_data)
self.logger.log_info(f"\n[RID: {request_id}] >>> [{'CHAT' if is_chat else 'TEXT'}][Input]: {log_input_str}")
# 2. Generation Config 생성
gen_config = self._create_generation_config(request)