Update handler.py

handler.py

@@ -100,6 +100,8 @@ class EndpointHandler():
         inputs = query.pop("inputs", query)
         if not inputs.endswith("<cite>"):
             inputs += "<cite>"
+        logger.info(inputs)
+        
         inputs = self.tokenizer(inputs, return_tensors="pt").to("cuda")
         with torch.no_grad():
             outputs = self.model(**inputs)