diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py
index 9fef920a9a..05b52e5439 100644
--- a/unsloth/models/llama.py
+++ b/unsloth/models/llama.py
@@ -203,9 +203,7 @@ def LlamaAttention_fast_forward_inference(
     A = torch.matmul(A, Vnn, out = Qn)
     A = A.transpose(1, 2)
     A = A.reshape(bsz, 1, attention_size)
-    print(self.temp_QA[1].shape)
-    A = fast_linear_forward(self.o_proj, A)#, out = self.temp_QA[1][:,:,self.hidden_size])
-    print(A.shape)
+    A = fast_linear_forward(self.o_proj, A, out = self.temp_QA[1][:,:,self.hidden_size])
     return A, (Kn, Vn)
 pass