Update app.py

app.py

@@ -85,7 +85,7 @@ def generate(
     
     retriever_inputs = retriever_tokenizer([message], max_length=1024, truncation=True, return_tensors="pt")
     with torch.no_grad():
-        embeddings = model(**retriever_inputs).pooler_output
+        embeddings = retriever(**retriever_inputs).pooler_output
         embeddings = embeddings.cpu().numpy()
 
     results = index.query(