diff --git a/tests/recipes/test_full_finetune.py b/tests/recipes/test_full_finetune.py
index b4f125080e..1f68291397 100644
--- a/tests/recipes/test_full_finetune.py
+++ b/tests/recipes/test_full_finetune.py
@@ -127,7 +127,7 @@ def test_loss(self, capsys, pytestconfig, tmpdir, monkeypatch):
                 --config {_CONFIG_PATH} \
                 output_dir={tmpdir} \
                 model=torchtune.models.{ckpt} \
-                checkpointer=torchtune.utils.{checkpointer}
+                checkpointer._component_=torchtune.utils.{checkpointer}
                 checkpointer.checkpoint_dir='{ckpt_dir}' \
                 checkpointer.checkpoint_files=[{ckpt_path}]\
                 checkpointer.model_type=LLAMA2
@@ -174,7 +174,7 @@ def test_training_state_on_resume(self, capsys, tmpdir, monkeypatch):
             --config {_CONFIG_PATH} \
             output_dir={ckpt_dir} \
             model=torchtune.models.{model_ckpt} \
-            checkpointer=torchtune.utils.FullModelHFCheckpointer \
+            checkpointer._component_=torchtune.utils.FullModelHFCheckpointer \
             checkpointer.checkpoint_dir='{ckpt_dir}' \
             checkpointer.checkpoint_files=[{ckpt_path}]\
             checkpointer.model_type=LLAMA2 \
@@ -194,7 +194,7 @@ def test_training_state_on_resume(self, capsys, tmpdir, monkeypatch):
             --config {_CONFIG_PATH} \
             output_dir={tmpdir} \
             model=torchtune.models.{model_ckpt} \
-            checkpointer=torchtune.utils.FullModelHFCheckpointer \
+            checkpointer._component_=torchtune.utils.FullModelHFCheckpointer \
             checkpointer.checkpoint_dir={ckpt_dir} \
             checkpointer.checkpoint_files=[{os.path.join(ckpt_dir, "hf_model_0001_2.pt")}]\
             checkpointer.model_type=LLAMA2 \
@@ -235,7 +235,7 @@ def test_gradient_accumulation(
         tune full_finetune \
             --config {_CONFIG_PATH} \
             model=torchtune.models.{model_ckpt} \
-            checkpointer=torchtune.utils.FullModelTorchTuneCheckpointer \
+            checkpointer._component_=torchtune.utils.FullModelTorchTuneCheckpointer \
             checkpointer.checkpoint_dir={ckpt_dir} \
             checkpointer.checkpoint_files=[{ckpt_path}]\
             checkpointer.model_type=LLAMA2 \
@@ -262,7 +262,7 @@ def test_gradient_accumulation(
         tune full_finetune \
             --config {_CONFIG_PATH} \
             model=torchtune.models.{model_ckpt} \
-            checkpointer=torchtune.utils.FullModelTorchTuneCheckpointer \
+            checkpointer._component_=torchtune.utils.FullModelTorchTuneCheckpointer \
             checkpointer.checkpoint_dir={ckpt_dir} \
             checkpointer.checkpoint_files=[{ckpt_path}]\
             checkpointer.model_type=LLAMA2 \
diff --git a/tests/recipes/test_lora_finetune.py b/tests/recipes/test_lora_finetune.py
index 0112c9d44b..6d4d22de01 100644
--- a/tests/recipes/test_lora_finetune.py
+++ b/tests/recipes/test_lora_finetune.py
@@ -118,9 +118,8 @@ def test_save_and_load_merged_weights(
         cmd = f"""
         tune {recipe_name}
             --config {config_path} \
-            --override \
             enable_fsdp={enable_fsdp} \
-            model._component_=torchtune.models.{ckpt} \
+            model=torchtune.models.{ckpt} \
             model_checkpoint={fetch_ckpt_model_path(ckpt)} \
             model.lora_rank=8 \
             model.lora_alpha=16 \