tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-26 11:13:53 -04:00
Code Issues Packages Projects Releases Wiki Activity

ggml : implement GEGLU_ERF and GEGLU_QUICK ops (#14445)

Browse Source
This commit is contained in:
Sigbjørn Skjæret
2025-07-03 23:07:22 +02:00
committed by GitHub
parent bee28421be
commit 28657a8229
20 changed files with 789 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
ggml/include/ggml.h
View File

@ -557,6 +557,8 @@ extern "C" {
GGML_GLU_OP_REGLU,
GGML_GLU_OP_GEGLU,
GGML_GLU_OP_SWIGLU,
GGML_GLU_OP_GEGLU_ERF,
GGML_GLU_OP_GEGLU_QUICK,
GGML_GLU_OP_COUNT,
};
@ -1147,6 +1149,22 @@ extern "C" {
struct ggml_context * ctx,
struct ggml_tensor * a);
GGML_API struct ggml_tensor * ggml_geglu_erf(
struct ggml_context * ctx,
struct ggml_tensor * a);
GGML_API struct ggml_tensor * ggml_geglu_erf_swapped(
struct ggml_context * ctx,
struct ggml_tensor * a);
GGML_API struct ggml_tensor * ggml_geglu_quick(
struct ggml_context * ctx,
struct ggml_tensor * a);
GGML_API struct ggml_tensor * ggml_geglu_quick_swapped(
struct ggml_context * ctx,
struct ggml_tensor * a);
// A: n columns, r rows,
// B: n columns, r rows,
GGML_API struct ggml_tensor * ggml_glu_split(
@ -1170,6 +1188,16 @@ extern "C" {
struct ggml_tensor * a,
struct ggml_tensor * b);
GGML_API struct ggml_tensor * ggml_geglu_erf_split(
struct ggml_context * ctx,
struct ggml_tensor * a,
struct ggml_tensor * b);
GGML_API struct ggml_tensor * ggml_geglu_quick_split(
struct ggml_context * ctx,
struct ggml_tensor * a,
struct ggml_tensor * b);
// normalize along rows
GGML_API struct ggml_tensor * ggml_norm(
struct ggml_context * ctx,