diff --git a/auto_gptq/modeling/llama.py b/auto_gptq/modeling/llama.py
index 52b94f1..6321199 100644
--- a/auto_gptq/modeling/llama.py
+++ b/auto_gptq/modeling/llama.py
@@ -12,10 +12,5 @@ class LlamaGPTQForCausalLM(BaseGPTQForCausalLM):
         ["mlp.down_proj"]
     ]
 
-    @staticmethod
-    def _resize_attention_mask(attention_mask):
-        attention_mask = [each.unsqueeze(1) for each in attention_mask]
-        return attention_mask
-
 
 __all__ = ["LlamaGPTQForCausalLM"]
diff --git a/auto_gptq/modeling/opt.py b/auto_gptq/modeling/opt.py
index 720765f..89066c2 100644
--- a/auto_gptq/modeling/opt.py
+++ b/auto_gptq/modeling/opt.py
@@ -15,10 +15,5 @@ class OPTGPTQForCausalLM(BaseGPTQForCausalLM):
         ["fc2"]
     ]
 
-    @staticmethod
-    def _resize_attention_mask(attention_mask):
-        attention_mask = [each.unsqueeze(1) for each in attention_mask]
-        return attention_mask
-
 
 __all__ = ["OPTGPTQForCausalLM"]