描述一下Bug
在kaggle上部署

`!git clone https://github.com/SakuraLLM/Sakura-13B-Galgame.git
%cd Sakura-13B-Galgame

!pip install llama-cpp-python --prefer-binary --extra-index-url=https://jllllll.github.io/llama-cpp-python-cuBLAS-wheels/AVX2/cu121
!pip install -q -r requirements.llamacpp.txt
!pip install -q pyngrok`

`ngrokToken = "************************************"
MODEL = "sakura-13b-lnovel-v0.9b-Q6_K"

from huggingface_hub import hf_hub_download
from pathlib import Path

if ngrokToken:
from pyngrok import conf,ngrok
conf.get_default().auth_token = ngrokToken
conf.get_default().monitor_thread = False
ssh_tunnels = ngrok.get_tunnels(conf.get_default())
if len(ssh_tunnels) == 0:
ssh_tunnel = ngrok.connect(5000)
print('address:' + ssh_tunnel.public_url)
else:
print('address:' + ssh_tunnels[0].public_url)

MODEL_PATH = f"./models/{MODEL}.gguf"
if not Path(MODEL_PATH).exists():
hf_hub_download(repo_id="SakuraLLM/Sakura-14B-LNovel-v0.9b-GGUF",filename = f"{MODEL}.gguf", local_dir = "models/")

!python server.py
--model_name_or_path $MODEL_PATH
--llama_cpp
--use_gpu
--model_version v0.9
--trust_remote_code
--no-auth`

代码如上。在运行基于qwen的0.9模型时正常运行。但换成基于qwen1.5模型的无论是0.9还是0.10pre，都会在中间显示qwen2是unknown，最后"assert self.model is not None"报错结束运行

python 3.10
llama-cpp-python 0.2.26
T4*2