diff --git a/keras_nlp/src/models/causal_lm.py b/keras_nlp/src/models/causal_lm.py
index 3493532e2c..c24150041e 100644
--- a/keras_nlp/src/models/causal_lm.py
+++ b/keras_nlp/src/models/causal_lm.py
@@ -352,9 +352,9 @@ def generate(
             raise ValueError(
                 'A `preprocessor` must be attached to the model if `stop_token_ids="auto"`. '
                 "Currently `preprocessor=None`. To call `generate()` with preprocessing "
-                "detached, either pass `stop_tokens_ids=None` to always generate until "
+                "detached, either pass `stop_token_ids=None` to always generate until "
                 "`max_length` or pass a tuple of token ids that should terminate generation "
-                "as `stop_tokens_ids`."
+                "as `stop_token_ids`."
             )
         elif stop_token_ids == "auto":
             stop_token_ids = [self.preprocessor.tokenizer.end_token_id]
diff --git a/keras_nlp/src/models/phi3/phi3_presets.py b/keras_nlp/src/models/phi3/phi3_presets.py
index 48ea0c1994..0f935d3371 100644
--- a/keras_nlp/src/models/phi3/phi3_presets.py
+++ b/keras_nlp/src/models/phi3/phi3_presets.py
@@ -22,7 +22,7 @@
                 "model. The model was trained using the Phi-3 datasets. This "
                 "dataset includes both synthetic data and filtered publicly "
                 "available website data, with an emphasis on high-quality and "
-                "reasoning-dense properties.",
+                "reasoning-dense properties."
             ),
             "params": 3821079552,
             "official_name": "Phi-3",
diff --git a/tools/checkpoint_conversion/convert_llama_checkpoints.py b/tools/checkpoint_conversion/convert_llama_checkpoints.py
index 27be78901b..b91dd51d1b 100644
--- a/tools/checkpoint_conversion/convert_llama_checkpoints.py
+++ b/tools/checkpoint_conversion/convert_llama_checkpoints.py
@@ -257,7 +257,7 @@ def main(_):
     test_model(keras_nlp_model, keras_nlp_tokenizer, hf_model, hf_tokenizer)
     print("\n-> Tests passed!")
 
-    keras_nlp_model.save_to_preset(keras_nlp_model, preset)
+    keras_nlp_model.save_to_preset(preset)
     print("\n-> Saved the model preset in float16")
 
     # === Save the tokenizer ===