update

models/cpp_qwen2.py

@@ -24,6 +24,8 @@ python convert_hf_to_gguf.py /workspace/xusong/huggingface/models/Qwen1.5-0.5B-C
 
 import json
 import copy
+import os
+
 from models.base_model import Simulator
 import llama_cpp
 # import llama_cpp.llama_tokenizer
@@ -57,7 +59,9 @@ class Qwen2Simulator(Simulator):
                 use_mlock=True,
                 verbose=False,
             )
-        logger.info(f"llm has been initialized: {self.llm}, n_threads={self.llm.n_threads}, n_ctx={self.llm.n_ctx}")
+        logger.info(f"llm has been initialized: {self.llm}, "
+                    f"n_threads={self.llm.n_threads}, n_ctx={self.llm.n_ctx}, "
+                    f"env[CACHE]={os.environ.get('CACHE', None)}")
 
         self.generation_kwargs = dict(
             temperature=config.DEFAULT_TEMPERATURE,