From 7355ca84b547bc87c3466dfbc485bee9ad2af735 Mon Sep 17 00:00:00 2001
From: simonJJJ <821898965@qq.com>
Date: Tue, 16 Apr 2024 13:29:42 +0800
Subject: [PATCH] fix-review

---
 llama.cpp | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/llama.cpp b/llama.cpp
index 14048d726..de0321919 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -8537,12 +8537,6 @@ struct llm_build_context {
                 Vcur = ggml_add(ctx0, Vcur, model.layers[il].bv);
                 cb(Vcur, "Vcur", il);
 
-                // these nodes are added to the graph together so that they are not reordered
-                // by doing so, the number of splits in the graph is reduced
-                ggml_build_forward_expand(gf, Qcur);
-                ggml_build_forward_expand(gf, Kcur);
-                ggml_build_forward_expand(gf, Vcur);
-
                 Qcur = ggml_rope_custom(
                     ctx0, ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens), inp_pos,
                     n_rot, rope_type, 0, n_orig_ctx, freq_base, freq_scale,