Update app.py

app.py

@@ -36,25 +36,22 @@ device = "cuda" if torch.cuda.is_available() else "cpu"
 #tokenizer = LlamaTokenizer.from_pretrained(REPO_NAME, use_auth_token=HF_TOKEN)
 
 #tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-tokenizer = AutoTokenizer.from_pretrained(
-    "microsoft/Phi-3-mini-4k-instruct",
-    trust_remote_code=True,
-    use_auth_token=HF_TOKEN
-)
+# tokenizer = AutoTokenizer.from_pretrained(
+#     "microsoft/Phi-3-mini-4k-instruct",
+#     trust_remote_code=True,
+#     use_auth_token=HF_TOKEN
+# )
+
+config = AutoConfig.from_pretrained(REPO_NAME, use_auth_token=HF_TOKEN)
+tokenizer = AutoTokenizer.from_pretrained(REPO_NAME, use_auth_token=HF_TOKEN)
 
 model = AutoModelForCausalLM.from_pretrained(
     REPO_NAME,
+    config = config,
     use_auth_token=HF_TOKEN,
     torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-    device_map="auto",
-    trust_remote_code=True
+    device_map="auto"
 )
-# model = AutoModelForCausalLM.from_pretrained(
-#     REPO_NAME,
-#     use_auth_token=HF_TOKEN,
-#     torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-#     device_map="auto"
-# )
 
 # Message and request models
 class ChatMessage(BaseModel):