diff --git a/unsloth/tokenizer_utils.py b/unsloth/tokenizer_utils.py
index b8f710b2c5..89e62c717a 100644
--- a/unsloth/tokenizer_utils.py
+++ b/unsloth/tokenizer_utils.py
@@ -916,7 +916,7 @@ def fix_untrained_tokens(model, tokenizer, train_dataset, eps = 1e-16):
 
     if bad_not_trainable:
         raise ValueError(
-            'Unsloth: Untrained tokens found, but embed_tokens & lm_head not trainable, causing NaNs. '\
+            f'Unsloth: Untrained tokens for [{where_untrained_set}] found, but embed_tokens & lm_head not trainable, causing NaNs. '\
             'Restart then add `embed_tokens` & `lm_head` to '\
             '`FastLanguageModel.get_peft_model(target_modules = [..., "embed_tokens", "lm_head",]). `'\
             'Are you using the `base` model? Instead, use the `instruct` version to silence this warning.',