diff --git a/nemo/collections/asr/parts/submodules/adapters/multi_head_attention_adapter_module.py b/nemo/collections/asr/parts/submodules/adapters/multi_head_attention_adapter_module.py
index d244de33c4f4..4f5f7364171e 100644
--- a/nemo/collections/asr/parts/submodules/adapters/multi_head_attention_adapter_module.py
+++ b/nemo/collections/asr/parts/submodules/adapters/multi_head_attention_adapter_module.py
@@ -143,7 +143,7 @@ def __init__(
             dropout_rate=dropout_rate,
             max_cache_len=0,
             use_pytorch_sdpa=use_pytorch_sdpa,
-            use_pytorch_sdpa_backends=use_pytorch_sdpa_backends
+            use_pytorch_sdpa_backends=use_pytorch_sdpa_backends,
         )
 
         self.pre_norm = nn.LayerNorm(n_feat)
diff --git a/nemo/collections/asr/parts/submodules/multi_head_attention.py b/nemo/collections/asr/parts/submodules/multi_head_attention.py
index 1d9e2c6a67e0..3cde743da833 100644
--- a/nemo/collections/asr/parts/submodules/multi_head_attention.py
+++ b/nemo/collections/asr/parts/submodules/multi_head_attention.py
@@ -167,7 +167,9 @@ def forward(self, query, key, value, mask, pos_emb=None, cache=None):
 
                 dropout_rate = self.dropout_rate if self.training else 0
                 with sdpa_kernel(self.use_pytorch_sdpa_backends):
-                    out = torch.nn.functional.scaled_dot_product_attention(q, k, v, attn_mask=mask, dropout_p=dropout_rate)
+                    out = torch.nn.functional.scaled_dot_product_attention(
+                        q, k, v, attn_mask=mask, dropout_p=dropout_rate
+                    )
 
                 # this IF block can be deleted when https://github.com/pytorch/pytorch/pull/131863 is in the stable version
                 if mask is not None: