From 133dd7a7373a3c9b5b0739497ecd266cfb6f70ae Mon Sep 17 00:00:00 2001
From: Casper <casperbh.96@gmail.com>
Date: Thu, 21 Sep 2023 13:30:57 +0200
Subject: [PATCH] Fix usage of engine

---
 awq/modules/fused/attn.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/awq/modules/fused/attn.py b/awq/modules/fused/attn.py
index 332b7163..73bedd26 100644
--- a/awq/modules/fused/attn.py
+++ b/awq/modules/fused/attn.py
@@ -219,7 +219,7 @@ def forward(
             xv = xv.view((bsz,) + self.attention_shapes["single_xv_view"])
 
             past_key_value = (xk, xv) if use_cache else None
-            attention_weight = awq_inference_engine.single_query_attention(
+            attention_weight = ft_inference_engine.single_query_attention(
                 xq, # query
                 xk, # key
                 xv, # value