Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 06:39:25 +01:00
Code Issues Packages Projects Releases Wiki Activity

iq1_s: fix dequantize on the CPU

Browse Source
This commit is contained in:
Iwan Kawrakow 2024-03-11 14:17:28 +01:00
parent 436c65e1a8
commit 5440a127c7
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml-quants.c
View File

@ -3456,7 +3456,7 @@ void dequantize_row_iq1_s(const block_iq1_s * restrict x, float * restrict y, in
const uint16_t * qh = x[i].qh; const uint16_t * qh = x[i].qh;
for (int ib = 0; ib < QK_K/32; ++ib) { for (int ib = 0; ib < QK_K/32; ++ib) {
const float dl = d * ((2*(qh[ib] >> 12) & 7) + 1); const float dl = d * (2*((qh[ib] >> 12) & 7) + 1);
const float delta = qh[ib] & 0x8000 ? -IQ1S_DELTA : IQ1S_DELTA; const float delta = qh[ib] & 0x8000 ? -IQ1S_DELTA : IQ1S_DELTA;
for (int l = 0; l < 4; ++l) { for (int l = 0; l < 4; ++l) {
const int8_t * grid = (const int8_t *)(iq1s_grid + (qs[l] | (((qh[ib] >> 3*l) & 7) << 8))); const int8_t * grid = (const int8_t *)(iq1s_grid + (qs[l] | (((qh[ib] >> 3*l) & 7) << 8)));