diff --git a/README.md b/README.md
index a138d646..826512c6 100644
--- a/README.md
+++ b/README.md
@@ -351,10 +351,9 @@ To utilize Ascend NPU devices for (distributed) training and inference, you need
 | torch-npu    | 2.2.0   | 2.2.0     |
 | deepspeed    | 0.13.2  | 0.13.2    |
 
-> [!NOTE]
-> Remember to use `ASCEND_RT_VISIBLE_DEVICES` instead of `CUDA_VISIBLE_DEVICES` to specify the device to use.
->
-> If you cannot infer model on NPU devices, try setting `do_sample: false` in the configurations.
+Remember to use `ASCEND_RT_VISIBLE_DEVICES` instead of `CUDA_VISIBLE_DEVICES` to specify the device to use.
+
+If you cannot infer model on NPU devices, try setting `do_sample: false` in the configurations.
 
 </details>
 
diff --git a/README_zh.md b/README_zh.md
index a0373711..d41ff13a 100644
--- a/README_zh.md
+++ b/README_zh.md
@@ -351,10 +351,9 @@ pip install https://github.com/jllllll/bitsandbytes-windows-webui/releases/downl
 | torch-npu    | 2.2.0   | 2.2.0     |
 | deepspeed    | 0.13.2  | 0.13.2    |
 
-> [!NOTE]
-> 请记得使用 `ASCEND_RT_VISIBLE_DEVICES` 而非 `CUDA_VISIBLE_DEVICES` 来指定您使用的设备。
->
-> 如果遇到无法正常推理的情况，请尝试设置 `do_sample: false`。
+请记得使用 `ASCEND_RT_VISIBLE_DEVICES` 而非 `CUDA_VISIBLE_DEVICES` 来指定您使用的设备。
+
+如果遇到无法正常推理的情况，请尝试设置 `do_sample: false`。
 
 </details>