diff --git a/docs/source/en/deepspeed.md b/docs/source/en/deepspeed.md
index ae304404f615..069e3835321e 100644
--- a/docs/source/en/deepspeed.md
+++ b/docs/source/en/deepspeed.md
@@ -196,7 +196,7 @@ ZeRO-2 shards the optimizer and gradient states across GPUs. This stage is prima
         "overlap_comm": true,
         "reduce_scatter": true,
         "reduce_bucket_size": 5e8,
-        "contiguous_gradients": true
+        "contiguous_gradients": true,
         "round_robin_gradients": true
     }
 }
diff --git a/docs/source/ko/deepspeed.md b/docs/source/ko/deepspeed.md
index 0390f65197ec..8c14905329e5 100644
--- a/docs/source/ko/deepspeed.md
+++ b/docs/source/ko/deepspeed.md
@@ -211,7 +211,7 @@ ZeRO-2는 GPU에서 옵티마이저와 그레이디언트를 분할합니다. 
         "overlap_comm": true,
         "reduce_scatter": true,
         "reduce_bucket_size": 5e8,
-        "contiguous_gradients": true
+        "contiguous_gradients": true,
         "round_robin_gradients": true
     }
 }