Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True (#29024)

* Adding _tie_weights() to prediction heads to support low_cpu_mem_usage=True

* Testing for the non-safe-tensors case, since the default is safe-tensors already

* Running fixup/fix-copies

* Adding accelerate annotations to tests

tests/models/sew_d/test_modeling_sew_d.py

@@ -460,6 +460,18 @@ class SEWDModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
     def test_feed_forward_chunking(self):
         pass
 
+    @unittest.skip("No support for low_cpu_mem_usage=True.")
+    def test_save_load_low_cpu_mem_usage(self):
+        pass
+
+    @unittest.skip("No support for low_cpu_mem_usage=True.")
+    def test_save_load_low_cpu_mem_usage_checkpoints(self):
+        pass
+
+    @unittest.skip("No support for low_cpu_mem_usage=True.")
+    def test_save_load_low_cpu_mem_usage_no_safetensors(self):
+        pass
+
     @slow
     def test_model_from_pretrained(self):
         model = SEWDModel.from_pretrained("asapp/sew-d-tiny-100k")