diff --git a/vllm_ascend/ops/vocab_parallel_embedding.py b/vllm_ascend/ops/vocab_parallel_embedding.py
index fe7ee5186c..0a7d7efb4f 100644
--- a/vllm_ascend/ops/vocab_parallel_embedding.py
+++ b/vllm_ascend/ops/vocab_parallel_embedding.py
@@ -253,16 +253,3 @@ def _get_logits_normal(
             logits = logits[..., :self.org_vocab_size]
 
         return logits
-
-    def forward(
-        self,
-        lm_head: VocabParallelEmbedding,
-        hidden_states: torch.Tensor,
-        # keep this for version compatibility
-        sampling_metadata=None,  # type: ignore
-        embedding_bias: Optional[torch.Tensor] = None,
-    ) -> Optional[torch.Tensor]:
-        return LogitsProcessor.forward(self,
-                                       lm_head,
-                                       hidden_states,
-                                       embedding_bias=embedding_bias)