updated max_new_tokens in handler.py

handler.py

@@ -54,7 +54,7 @@ class EndpointHandler:
                 outputs = self.model.generate(
                     tokenizer_output["input_ids"],
                     attention_mask=tokenizer_output["attention_mask"],
-                    max_length=512,
+                    max_new_tokens=1024,
                     pad_token_id=self.tokenizer.pad_token_id,
                     do_sample=True,
                     temperature=0.7,