tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 12:05:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: enable Gemma FA for HIP/Pascal (#9581)

Browse Source
This commit is contained in:
Johannes Gäßler
2024-09-22 09:34:52 +02:00
committed by GitHub
parent ecd5d6b65b
commit a5b57b08ce
3 changed files with 12 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/test-backend-ops.cpp
View File

@ -3599,7 +3599,7 @@ static bool test_backend(ggml_backend_t backend, test_mode mode, const char * op
if (hs != 128 && logit_softcap != 0.0f) continue;
for (int nh : { 32, }) {
for (int kv : { 512, 1024, }) {
for (int nb : { 1, 2, 4, 8, }) {
for (int nb : { 1, 3, 32, 35, }) {
for (ggml_type type_KV : {GGML_TYPE_F16, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0}) {
test_cases.emplace_back(new test_flash_attn_ext(hs, nh, kv, nb, mask, max_bias, logit_softcap, type_KV));
}