diff --git a/litellm/proxy/spend_tracking/spend_tracking_utils.py b/litellm/proxy/spend_tracking/spend_tracking_utils.py
index ebb7a5ba1d7..e69c0c80420 100644
--- a/litellm/proxy/spend_tracking/spend_tracking_utils.py
+++ b/litellm/proxy/spend_tracking/spend_tracking_utils.py
@@ -1095,16 +1095,14 @@ def _sanitize_request_body_for_spend_logs_payload(
     request_body: Mapping[str, object],
     visited: set | None = None,
     max_string_length_prompt_in_db: int | None = None,
-    redact_credentials: bool = False,
 ) -> dict:
     """
     Recursively sanitize request body to prevent logging large base64 strings or other large values.
     Truncates strings longer than MAX_STRING_LENGTH_PROMPT_IN_DB characters and handles nested dictionaries.
 
     At every nesting level, also strips keys listed in _SENSITIVE_REQUEST_BODY_KEYS (e.g. secret_fields,
-    which holds raw HTTP headers including Authorization tokens). With ``redact_credentials``, string
-    values under keys SensitiveDataMasker classifies as credentials are replaced with
-    REDACTED_BY_LITELM_STRING.
+    which holds raw HTTP headers including Authorization tokens), and replaces string values under keys
+    SensitiveDataMasker classifies as credentials with REDACTED_BY_LITELM_STRING.
     """
     from litellm.constants import (
         LITELLM_TRUNCATED_PAYLOAD_FIELD,
@@ -1124,9 +1122,7 @@ def _sanitize_request_body_for_spend_logs_payload(
 
     def _sanitize_value(value: object) -> object:
         if isinstance(value, Mapping):
-            return _sanitize_request_body_for_spend_logs_payload(
-                value, visited, max_string_length_prompt_in_db, redact_credentials
-            )
+            return _sanitize_request_body_for_spend_logs_payload(value, visited, max_string_length_prompt_in_db)
         elif isinstance(value, list):
             return [_sanitize_value(item) for item in value]
         elif isinstance(value, str):
@@ -1164,7 +1160,7 @@ def _sanitize_request_body_for_spend_logs_payload(
         return value
 
     return {
-        k: REDACTED_BY_LITELM_STRING if redact_credentials and _is_request_body_credential(k, v) else _sanitize_value(v)
+        k: REDACTED_BY_LITELM_STRING if _is_request_body_credential(k, v) else _sanitize_value(v)
         for k, v in request_body.items()
         if k not in _SENSITIVE_REQUEST_BODY_KEYS
     }
@@ -1582,7 +1578,7 @@ def _get_proxy_server_request_for_spend_logs_payload(
                     _request_body = _convert_mapping_to_json_serializable(without_classifier_audit(_request_body))
                     perform_redaction(model_call_details=_request_body, result=None)
 
-            _request_body = _sanitize_request_body_for_spend_logs_payload(_request_body, redact_credentials=True)
+            _request_body = _sanitize_request_body_for_spend_logs_payload(_request_body)
             _request_body_json_str: Final = safe_dumps(_request_body)
             if LITELLM_TRUNCATED_PAYLOAD_FIELD in _request_body_json_str:
                 verbose_proxy_logger.info(
diff --git a/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py b/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py
index 9128c384b8c..f3991c0e494 100644
--- a/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py
+++ b/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py
@@ -612,7 +612,7 @@ def test_sanitize_request_body_for_spend_logs_payload_mixed_types():
     request_body = {
         "text": long_string,
         "number": 42,
-        "nested": {"list": ["short", long_string], "dict": {"key": long_string}},
+        "nested": {"list": ["short", long_string], "dict": {"value": long_string}},
     }
     sanitized = _sanitize_request_body_for_spend_logs_payload(request_body)
 
@@ -631,7 +631,7 @@ def test_sanitize_request_body_for_spend_logs_payload_mixed_types():
     assert sanitized["number"] == 42
     assert sanitized["nested"]["list"][0] == "short"
     assert len(sanitized["nested"]["list"][1]) == expected_length
-    assert len(sanitized["nested"]["dict"]["key"]) == expected_length
+    assert len(sanitized["nested"]["dict"]["value"]) == expected_length
 
 
 def test_sanitize_request_body_for_spend_logs_payload_uses_runtime_env_override(
@@ -2781,10 +2781,12 @@ def test_proxy_server_request_payload_redacts_provider_credentials(mock_should_s
     assert parsed["messages"] == [{"role": "user", "content": "hello"}]
 
 
-def test_sanitize_request_body_keeps_credential_named_fields_by_default() -> None:
-    response: Final = {"system_fingerprint": "fp_123", "usage": {"prompt_tokens": 1}}
+def test_sanitize_response_redacts_credential_named_fields() -> None:
+    response: Final = {"access_token": "canary-oauth-token", "usage": {"prompt_tokens": 1}}
 
-    assert _sanitize_request_body_for_spend_logs_payload({"response": response}) == {"response": response}
+    assert _sanitize_request_body_for_spend_logs_payload({"response": response}) == {
+        "response": {"access_token": REDACTED_BY_LITELM_STRING, "usage": {"prompt_tokens": 1}}
+    }
 
 
 @patch("litellm.proxy.spend_tracking.spend_tracking_utils.should_store_prompts_and_responses_in_spend_logs")