guard size mismatch check to only quantized models (#38397)

fix

src/transformers/modeling_utils.py

@@ -1577,7 +1577,8 @@ def _find_mismatched_keys(
                 # This skips size mismatches for 4-bit weights. Two 4-bit values share an 8-bit container, causing size differences.
                 # Without matching with module type or parameter type it seems like a practical way to detect valid 4bit weights.
                 if not (
-                    new_state_dict[key].shape[-1] == 1
+                    is_quantized
+                    and new_state_dict[key].shape[-1] == 1
                     and new_state_dict[key].numel() * 2 == model_state_dict[key].numel()
                 ):
                     mismatched_keys.append(key)