diff --git a/convert-falcon-hf-to-gguf.py b/convert-falcon-hf-to-gguf.py
index b3e190a0f..32420796b 100644
--- a/convert-falcon-hf-to-gguf.py
+++ b/convert-falcon-hf-to-gguf.py
@@ -109,6 +109,8 @@ gguf_writer.add_layer_norm_eps(hparams["layer_norm_epsilon"])
 print("gguf: get tokenizer metadata")
 
 tokens: List[str] = []
+scores: List[float] = []
+toktypes: List[int] = []
 merges: List[str] = []
 
 
@@ -152,8 +154,12 @@ if Path(dir_model + "/tokenizer.json").is_file():
             text = bytearray(pad_token)
 
         tokens.append(text)
+        scores.append(0.0)                      # dymmy
+        toktypes.append(gguf.TokenType.NORMAL)  # dummy
 
     gguf_writer.add_token_list(tokens)
+    gguf_writer.add_token_scores(scores)
+    gguf_writer.add_token_types(toktypes)
 
     if "added_tokens" in tokenizer_json and Path(dir_model + "/tokenizer_config.json").is_file():
         print("gguf: get special token ids")
diff --git a/llama.cpp b/llama.cpp
index 5c0bf6190..61e989810 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -1968,7 +1968,7 @@ static bool llama_model_load(
                 } break;
             case LLM_ARCH_FALCON:
                 {
-                }
+                } break;
             default:
                 throw std::runtime_error("unsupported architecture");
         };