Update app.py

app.py

@@ -58,7 +58,7 @@ cls_model_name = "saiteki-kai/QA-DeBERTa-v3-large-binary-3"
 
 model = transformers.pipeline(
     model=chat_model_name,
-    model_kwargs={"dtype": torch.bfloat16} | ({"attn_implementation": "flash_attention_2"} if is_flash_attn_2_available() else {}),
+    model_kwargs={"dtype": torch.bfloat16, "padding_side": "left"} | ({"attn_implementation": "flash_attention_2"} if is_flash_attn_2_available() else {}),
     device=device,
 )