Update README.md

README.md

@@ -282,9 +282,9 @@ class LLM:
               add_generation_prompt=True, # Must add for generation
               return_tensors="pt",
         ).to("cuda")
-
+        
         # Increase the max_new_tokens to allow more detailed responses
-        output_ids = self.model.generate(inputs, max_new_tokens=self.max_new_tokens, pad_token_id=self._EOS_TOKEN_ID)
+        output_ids = self.model.generate(inputs, max_new_tokens=self.max_new_tokens, pad_token_id=self.__tokenizer.pad_token_id)
         output_ids = output_ids.tolist()[0] if output_ids.size(0) == 1 else output_ids.tolist()
 
         output_text = self.__tokenizer.decode(output_ids, skip_special_tokens=True)
@@ -320,4 +320,7 @@ class LLM:
             },
             "methods": ["invoke", "extract_answer"]
         }
+
+
+llm = LLM()
 ```