diff --git a/CHANGELOG.md b/CHANGELOG.md
index 058c1a6..a089397 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,11 +6,17 @@
   `SimulatorDevice` class for general IonQ devices.
   [(#50)](https://github.com/PennyLaneAI/PennyLane-IonQ/pull/50)
 
+### Bug Fixes
+
+* Since the histogram of probabilities returned from the remote simulator does not always sum exactly to one,
+  the PennyLane device normalizes them to higher precision.
+  [(#53)](https://github.com/PennyLaneAI/PennyLane-IonQ/pull/53)
+
 ### Contributors
 
 This release contains contributions from (in alphabetical order):
 
-Jon Donovan, Antal Száva
+Jon Donovan, Christina Lee, Antal Száva
 
 ---
 
diff --git a/pennylane_ionq/device.py b/pennylane_ionq/device.py
index 1c4fd69..596edc0 100644
--- a/pennylane_ionq/device.py
+++ b/pennylane_ionq/device.py
@@ -129,7 +129,6 @@ def apply(self, operations, **kwargs):
         for operation in rotations:
             self._apply_operation(operation)
 
-        # print(self.job)
         self._submit_job()
 
     def _apply_operation(self, operation):
@@ -194,7 +193,11 @@ def prob(self):
 
             # convert the sparse probs into a probability array
             self._prob_array = np.zeros([2**self.num_wires])
-            self._prob_array[idx] = np.fromiter(self.histogram.values(), float)
+
+            # histogram values don't always perfectly sum to exactly one
+            histogram_values = self.histogram.values()
+            norm = sum(histogram_values)
+            self._prob_array[idx] = np.fromiter(histogram_values, float) / norm
 
         return self._prob_array
 
diff --git a/tests/conftest.py b/tests/conftest.py
index 1ba979b..b9e7901 100755
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -79,7 +79,7 @@ def init_state(scope="session"):
 
     def _init_state(n):
         state = np.random.randint(0, 2, [n])
-        ket = np.zeros([2 ** n])
+        ket = np.zeros([2**n])
         ket[np.ravel_multi_index(state, [2] * n)] = 1
         return state, ket