vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Merge branch 'master' into gg/flash-attn

Browse source
This commit is contained in:
Georgi Gerganov 2024-04-17 10:13:09 +03:00
commit 2c41180e88
No known key found for this signature in database
GPG key ID: BF970631944C16B7
110 changed files with 11660 additions and 6357 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda/common.cuh
View file

@ -393,7 +393,7 @@ static __device__ __forceinline__ int __dp4a(const int a, const int b, int c) {
// TODO: move to ggml-common.h
static const __device__ int8_t kvalues_iq4nl[16] = {-127, -104, -83, -65, -49, -35, -22, -10, 1, 13, 25, 38, 53, 69, 89, 113};
typedef void (*dequantize_kernel_t)(const void * vx, const int ib, const int iqs, dfloat2 & v);
typedef void (*dequantize_kernel_t)(const void * vx, const int64_t ib, const int iqs, dfloat2 & v);
//////////////////////