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

iq1s_blocks16: Use 2*<x^2> as sigma2 in weight adjustment

Browse source
This commit is contained in:
Iwan Kawrakow 2024-03-08 15:14:28 +02:00
parent 4c4404ace5
commit c55e66f997
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-quants.c
View file

@ -11606,7 +11606,7 @@ static void quantize_row_iq1_s_impl(const float * restrict x, void * restrict vy
const float * xbl = x + QK_K*ibl; const float * xbl = x + QK_K*ibl;
float sumx2 = 0; float sumx2 = 0;
for (int i = 0; i < QK_K; ++i) sumx2 += xbl[i]*xbl[i]; for (int i = 0; i < QK_K; ++i) sumx2 += xbl[i]*xbl[i];
float sigma2 = sumx2/QK_K; float sigma2 = 2*sumx2/QK_K;
for (int ib = 0; ib < QK_K/IQ1S_BLOCK_SIZE; ++ib) { for (int ib = 0; ib < QK_K/IQ1S_BLOCK_SIZE; ++ib) {
const float * xb = xbl + IQ1S_BLOCK_SIZE*ib; const float * xb = xbl + IQ1S_BLOCK_SIZE*ib;