vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-16 20:09:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix GPT2 not loading due to graph too small

Browse source
This commit is contained in:
Concedo 2023-11-26 23:06:42 +08:00
parent eb42c73953
commit a6eb9b8010
8 changed files with 21 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

2
otherarch/llama_v3.cpp
View file

@ -12,6 +12,7 @@
#include "llama_v3.h"
#include "ggml.h"
#include "otherarch.h"
#ifdef GGML_USE_CUBLAS
#include "ggml-cuda.h"
#endif
@ -88,7 +89,6 @@ enum e_model3 {
static const size_t kB3 = 1024;
static const size_t MB3 = 1024*1024;
static const size_t GGML_MAX_NODES = 8192;
// computed for n_ctx == 2048
// TODO: dynamically determine these sizes