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

int64 dst

Browse source
This commit is contained in:
Georgi Gerganov 2024-11-07 17:17:29 +02:00
parent 3ab47eb746
commit 4abeb60a1a
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ggml/src/ggml-metal.metal
View file

@ -3234,10 +3234,10 @@ kernel void kernel_flash_attn_ext(
// final rescale with 1/S and store to global memory
if (sgitg == 0) {
for (short j = 0; j < Q && iq1 + j < ne01; ++j) {
const float S = ss[j*TS + 0];
const half S = ss[j*TS + 0];
for (short i = tiisg; i < D4; i += NW) {
dst4[(iq3*ne2*ne1 + iq2 + (iq1 + j)*ne1)*D4 + i] = (float4) so4[j*T4 + i]/S;
dst4[((int64_t)iq3*ne2*ne1 + iq2 + (iq1 + j)*ne1)*D4 + i] = (float4) so4[j*T4 + i]/S;
}
}
}