diff --git a/eval.py b/eval.py
index 76e3e5205add87c4a7e298f32177d3ffc37f481d..a9c426ec2c1c702b9a43ba7fb99f8e86068b596e 100644
--- a/eval.py
+++ b/eval.py
@@ -33,7 +33,7 @@ dataset2metric = {
     "lsht": classification_score,
     "passage_retrieval_en": retrieval_score,
     "passage_count": count_score,
-    "passkey_retrieval_zh": retrieval_zh_score,
+    "passage_retrieval_zh": retrieval_zh_score,
     "lcc": code_sim_score,
     "repobench-p": code_sim_score,
 }