Update app.py

app.py

@@ -89,8 +89,17 @@ def main():
 
         # loading remote llama model
         #llm = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf")
-        llm = AutoModelForCausalLM.from_pretrained("google/gemma-1.1-2b-it")
+        #llm = AutoModelForCausalLM.from_pretrained("google/gemma-1.1-2b-it")
 
+        token = os.environ["HF_TOKEN"]
+        llm = AutoModelForCausalLM.from_pretrained(
+        "google/gemma-7b-it",
+        # torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+        torch_dtype=torch.float16,
+        token=token,
+        )
+
+        
         # custom prompt
         custom_template="""
 You are a smart personal assistant and your task is to provide the answer of the given question based only on the given context. \n
@@ -111,7 +120,7 @@ Helpful Answer:
         qa = ConversationalRetrievalChain.from_llm(llm,
                                                    #chain_type = "stuff",
                                                    chain_type = "stuff",
-                                                   verbose=True,
+                                                   #verbose=True,
                                                    #retriever=docsearch.as_retriever()
                                                    retriever=docsearch.as_retriever(search_kwargs = {"k" : 4, "search_type" : "similarity"}),
                                                    combine_docs_chain_kwargs={"prompt": QA_PROMPT}