diff --git a/src/amplitude_experiment/local/client.py b/src/amplitude_experiment/local/client.py
index 2a73954..99c5d5c 100644
--- a/src/amplitude_experiment/local/client.py
+++ b/src/amplitude_experiment/local/client.py
@@ -106,7 +106,7 @@ def evaluate_v2(self, user: User, flag_keys: Set[str] = None, options: EvaluateO
         flag_configs = self.flag_config_storage.get_flag_configs()
         if flag_configs is None or len(flag_configs) == 0:
             return {}
-        self.logger.debug(f"[Experiment] Evaluate: user={user} - Flags: {flag_configs}")
+        self.logger.debug("[Experiment] Evaluate: user=%s - Flags: %s", user, flag_configs)
         sorted_flags = topological_sort(flag_configs, flag_keys and list(flag_keys))
         if not sorted_flags:
             return {}
@@ -126,7 +126,7 @@ def evaluate_v2(self, user: User, flag_keys: Set[str] = None, options: EvaluateO
                 metadata=v.metadata
             ) for k, v in result.items()
         }
-        self.logger.debug(f"[Experiment] Evaluate Result: {variants}")
+        self.logger.debug("[Experiment] Evaluate Result: %s", variants)
         if self.exposure_service is not None and options and options.tracks_exposure is True:
             self.exposure_service.track(Exposure(user, variants))
         if self.assignment_service is not None:
diff --git a/tests/local/evaluate_logging_test.py b/tests/local/evaluate_logging_test.py
new file mode 100644
index 0000000..40902f1
--- /dev/null
+++ b/tests/local/evaluate_logging_test.py
@@ -0,0 +1,55 @@
+import logging
+import unittest
+from unittest.mock import patch
+
+from src.amplitude_experiment import LocalEvaluationClient, LocalEvaluationConfig, User
+from src.amplitude_experiment.evaluation.types import EvaluationFlag
+
+API_KEY = 'server-api-key'
+
+
+def always_on_flag(key: str) -> EvaluationFlag:
+    return EvaluationFlag.from_dict({
+        'key': key,
+        'variants': {'on': {'key': 'on', 'value': 'on'}},
+        'segments': [{'variant': 'on'}],
+    })
+
+
+def forbidden_repr(self) -> str:
+    raise AssertionError('a flag config was rendered to text')
+
+
+class LocalEvaluationClientEvaluateLoggingTestCase(unittest.TestCase):
+
+    def _client(self, debug: bool) -> LocalEvaluationClient:
+        # Never started: no poller thread, no network call.
+        logger = logging.getLogger(f'evaluate-logging-test-{debug}')
+        logger.setLevel(logging.DEBUG if debug else logging.WARNING)
+        client = LocalEvaluationClient(API_KEY, LocalEvaluationConfig(logger=logger))
+        client.flag_config_storage.put_flag_config(always_on_flag('flag-a'))
+        client.flag_config_storage.put_flag_config(always_on_flag('flag-b'))
+        return client
+
+    def test_evaluate_does_not_render_flag_configs_when_debug_is_disabled(self):
+        client = self._client(debug=False)
+
+        with patch.object(EvaluationFlag, '__repr__', forbidden_repr):
+            variants = client.evaluate_v2(User(user_id='user'), {'flag-a'})
+
+        self.assertEqual('on', variants['flag-a'].value)
+
+    def test_evaluate_logs_flag_configs_and_result_when_debug_is_enabled(self):
+        client = self._client(debug=True)
+
+        with self.assertLogs(client.logger, logging.DEBUG) as logs:
+            client.evaluate_v2(User(user_id='user'), {'flag-a'})
+
+        self.assertEqual(2, len(logs.output))
+        self.assertIn('Evaluate: user={"device_id": null, "user_id": "user"', logs.output[0])
+        self.assertIn("EvaluationFlag(key='flag-b'", logs.output[0])
+        self.assertIn("Evaluate Result: {'flag-a': ", logs.output[1])
+
+
+if __name__ == '__main__':
+    unittest.main()