Sagemaker test fix (#10987)

* wrong makefile command

* ddp test fix

tests/sagemaker/README.md

@@ -18,7 +18,7 @@ git+https://github.com/huggingface/transformers.git@v4.5.0.rc0 # install master
 After we adjusted the `requirements.txt` we can run Amazon SageMaker tests with:  
 
 ```bash
-AWS_PROFILE=<enter-your-profile> make sagemaker-test
+AWS_PROFILE=<enter-your-profile> make test-sagemaker
 ```
 These tests take around 10-15 minutes to finish. Preferably make a screenshot of the successfully ran tests.
 
@@ -88,7 +88,7 @@ tensorflow-gpu==2.5.0 # for tensorflow
 After we adjusted the `requirements.txt` we can run Amazon SageMaker tests with. 
 
 ```bash
-AWS_PROFILE=<enter-your-profile> make sagemaker-test
+AWS_PROFILE=<enter-your-profile> make test-sagemaker
 ```
 These tests take around 10-15 minutes to finish. Preferably make a screenshot of the successfully ran tests.
 

tests/sagemaker/test_multi_node_data_parallel.py

@@ -58,7 +58,7 @@ class MultiNodeTest(unittest.TestCase):
     def create_estimator(self, instance_count):
         job_name = f"{self.env.base_job_name}-{instance_count}-{'ddp' if 'ddp' in self.script else 'smd'}"
         # distributed data settings
-        distribution = {"smdistributed": {"dataparallel": {"enabled": True}}}
+        distribution = {"smdistributed": {"dataparallel": {"enabled": True}}} if self.script != "run_ddp.py" else None
 
         # creates estimator
         return HuggingFace(