diff --git a/tensordict/tensordict.py b/tensordict/tensordict.py
index df0e2150b..8615704a7 100644
--- a/tensordict/tensordict.py
+++ b/tensordict/tensordict.py
@@ -2587,6 +2587,7 @@ def memmap_like(self, prefix: str | None = None) -> T:
                     prefix / f"{key}.meta.pt",
                 )
         tensordict._is_memmap = True
+        self._is_shared = False
         tensordict._device = torch.device("cpu")
         tensordict.lock_()
         return tensordict
@@ -4423,8 +4424,9 @@ def _set_str(
         inplace: bool,
         validated: bool,
     ) -> T:
-        best_attempt = inplace is BEST_ATTEMPT_INPLACE
-        inplace = self._convert_inplace(inplace, key)
+        if inplace is not False:
+            best_attempt = inplace is BEST_ATTEMPT_INPLACE
+            inplace = self._convert_inplace(inplace, key)
         if not validated:
             value = self._validate_value(value, check_shape=True)
         if not inplace:
@@ -4702,6 +4704,7 @@ def memmap_(
                     prefix / f"{key}.meta.pt",
                 )
         self._is_memmap = True
+        self._is_shared = False
         self._device = torch.device("cpu")
         self.lock_()
         return self
@@ -7707,6 +7710,7 @@ def memmap_(self, prefix: str | None = None, copy_existing: bool = False) -> T:
                 copy_existing=copy_existing,
             )
         self._is_memmap = True
+        self._is_shared = False
         self._device = torch.device("cpu")
         self.lock_()
         return self
@@ -7728,6 +7732,7 @@ def memmap_like(
             tds.append(td_like)
         td_out = torch.stack(tds, self.stack_dim)
         self._is_memmap = True
+        self._is_shared = False
         self._device = torch.device("cpu")
         td_out.lock_()
         return td_out
@@ -8447,6 +8452,7 @@ def memmap_(
             torch.save(metadata, prefix / "meta.pt")
 
         self._is_memmap = True
+        self._is_shared = False
         self._device = torch.device("cpu")
         self.lock_()
         return self
diff --git a/test/_utils_internal.py b/test/_utils_internal.py
index 669174a4e..aea83fa2b 100644
--- a/test/_utils_internal.py
+++ b/test/_utils_internal.py
@@ -31,6 +31,8 @@ def get_available_devices():
     if n_cuda > 0:
         for i in range(n_cuda):
             devices += [torch.device(f"cuda:{i}")]
+            if i == 1:
+                break
     return devices
 
 
diff --git a/test/test_memmap.py b/test/test_memmap.py
index 081309aba..4cef90d71 100644
--- a/test/test_memmap.py
+++ b/test/test_memmap.py
@@ -73,17 +73,7 @@ def test_memmap_same_device_as_tensor(device):
     """
     t = torch.tensor([1], device=device)
     m = MemoryMappedTensor.from_tensor(t)
-    assert m.device == torch.device(device)
-    for other_device in get_available_devices():
-        if other_device != device:
-            with pytest.raises(
-                RuntimeError,
-                match="Expected all tensors to be on the same device, "
-                + "but found at least two devices",
-            ):
-                assert torch.all(m + torch.ones([3, 4], device=other_device) == 1)
-        m = m.to(other_device)
-        assert m.device == torch.device(other_device)
+    assert m.device == torch.device("cpu")
 
 
 @pytest.mark.parametrize("device", get_available_devices())