diff --git a/torch/distributed/rpc/options.py b/torch/distributed/rpc/options.py
index 19092c283ef..64796b96d4e 100644
--- a/torch/distributed/rpc/options.py
+++ b/torch/distributed/rpc/options.py
@@ -73,16 +73,16 @@ class TensorPipeRpcBackendOptions(_TensorPipeRpcBackendOptionsBase):
             >>> # on worker 0
             >>> options = TensorPipeRpcBackendOptions(
             >>>     num_worker_threads=8,
-            >>>     device_maps={"worker1": {0, 1}}
+            >>>     device_maps={"worker1": {0: 1}}
             >>>     # maps worker0's cuda:0 to worker1's cuda:1
             >>> )
-            >>> options.set_device_map("worker1", {1, 2})
+            >>> options.set_device_map("worker1", {1: 2})
             >>> # maps worker0's cuda:1 to worker1's cuda:2
             >>>
             >>> rpc.init_rpc(
             >>>     "worker0",
             >>>     rank=0,
-            >>>     world_size=2
+            >>>     world_size=2,
             >>>     backend=rpc.BackendType.TENSORPIPE,
             >>>     rpc_backend_options=options
             >>> )
@@ -94,7 +94,7 @@ class TensorPipeRpcBackendOptions(_TensorPipeRpcBackendOptionsBase):
             >>> # the device map, and hence will be moved back to cuda:0 and
             >>> # cuda:1 on worker0
             >>> print(rets[0])  # tensor([2., 2.], device='cuda:0')
-            >>> print(rets[0])  # tensor([2., 2.], device='cuda:1')
+            >>> print(rets[1])  # tensor([2., 2.], device='cuda:1')
         """
         device_index_map = {}
         curr_device_maps = super().device_maps