fix: Little kv routing fix (#1677)

examples/llm/components/kv_router.py

@@ -100,8 +100,7 @@ def parse_args(service_name, prefix) -> Namespace:
     )
     parser.add_argument(
         "--softmax-sample",
-        type=bool,
-        default=False,
+        action="store_true",
         help="Whether to do softmax sampling based on worker logits (default is to pick smallest)",
     )
     config = ServiceConfig.get_instance()

examples/llm/configs/agg_router.yaml

@@ -29,7 +29,7 @@ Processor:
 
 Router:
   min-workers: 1
-  softmax_sample: true
+  softmax-sample: true
   common-configs: [model, block-size, router]
 
 VllmWorker: