zhangyiss/gguf-tools
1
0
Fork 0
mirror of https://github.com/antirez/gguf-tools.git synced 2025-09-16 17:48:08 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix q4_1 dequantization

Browse Source
This commit is contained in:
Juarez Bochi
2024-01-10 10:17:13 -05:00
parent fe34f6ec5c
commit 55d6267c31
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
gguflib.c
View File

@@ -834,7 +834,7 @@ void gguf_q4_1_to_float(void *weights_data, void *dst, uint64_t count, store_flo
/* For each block get the scale and convert all the
* weights in the block. */
float scale = from_half(*((uint16_t*)block));
float bias = from_half(*((uint16_t*)block+2));
float bias = from_half(*((uint16_t*)block+1));
/* First 16 weights are in the lower bits */
for (uint32_t j = 0; j < 16; j++) {
uint8_t value = block[j+4]; // j+2 to skip the scale and bias bytes.