423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-11-05 18:32:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix bug in vllm engine

Browse Source 
Former-commit-id: 38f02a2c5b52cba6908c2d3c2a455677f8574faf
This commit is contained in:
hiyouga 2024-05-15 02:17:54 +08:00
parent 200689e07c
commit ed6793afdf
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/llmtuner/chat/vllm_engine.py
View File

@ -100,8 +100,9 @@ class VllmEngine(BaseEngine):
max_new_tokens = input_kwargs.pop("max_new_tokens", None) max_new_tokens = input_kwargs.pop("max_new_tokens", None)
stop = input_kwargs.pop("stop", None) stop = input_kwargs.pop("stop", None)
max_tokens = self.generating_args["max_new_tokens"] or self.generating_args["max_length"]
if max_length: if max_length:
max_tokens = max_length - prompt_length max_tokens = max_length - prompt_length if max_length > prompt_length else 1
if max_new_tokens: if max_new_tokens:
max_tokens = max_new_tokens max_tokens = max_new_tokens