CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-01 09:01:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml-quants, llama : removed excess checks (#7274)

Browse Source
This commit is contained in:
Herman Semenov
2024-05-17 07:08:49 +00:00
committed by GitHub
parent e18bc6aaf3
commit 359cbe3f46
3 changed files with 4 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml-quants.c
View File

@@ -1986,7 +1986,7 @@ static void quantize_row_q3_K_impl(const float * restrict x, block_q3_K * restri
for (int j = 0; j < QK_K/16; ++j) {
if (quant_weights) {
const float * qw = quant_weights ? quant_weights + QK_K * i + 16*j : NULL;
const float * qw = quant_weights + QK_K * i + 16*j;
for (int l = 0; l < 16; ++l) weight[l] = qw[l] * sqrtf(sigma2 + x[16*j+l]*x[16*j+l]);
} else {
for (int l = 0; l < 16; ++l) weight[l] = x[16*j+l]*x[16*j+l];