CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-04 09:32:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update src/llama-model.cpp

Browse Source 
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Younes B
2025-07-08 12:30:55 +04:00
committed by GitHub
parent 52d1ef35ba
commit 58e3866d02
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama-model.cpp
View File

@@ -14868,7 +14868,7 @@ struct llm_build_falcon_h1 : public llm_graph_context {
// TODO: skip computing output earlier for unused tokens // TODO: skip computing output earlier for unused tokens
y = ggml_add(ctx0, y, ggml_mul(ctx0, x, model.layers[il].ssm_d)); y = ggml_add(ctx0, y, ggml_mul(ctx0, x, model.layers[il].ssm_d));
y = ggml_mul(ctx0, y, ggml_silu(ctx0, ggml_cont(ctx0, z))); y = ggml_swiglu_split(ctx0, ggml_cont(ctx0, z), y);
// grouped RMS norm // grouped RMS norm
if (model.layers[il].ssm_norm) { if (model.layers[il].ssm_norm) {