root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-30 13:24:35 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

tests : add hs=256

Browse Source
This commit is contained in:
Georgi Gerganov 2024-03-24 12:21:41 +02:00
parent 09532120e0
commit e425810bb6
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml.c
View File

@ -12272,7 +12272,7 @@ static void ggml_compute_forward_soft_max_f32(
GGML_TENSOR_UNARY_OP_LOCALS
const int64_t ne11 = src1 ? src1->ne[1] : 1;
//const int64_t ne11 = src1 ? src1->ne[1] : 1;
// TODO: is this supposed to be ceil instead of floor?
// https://huggingface.co/mosaicml/mpt-7b/blob/main/attention.py#L370

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

@ -2261,7 +2261,7 @@ static bool test_backend(ggml_backend_t backend, test_mode mode, const char * op
test_cases.emplace_back(new test_leaky_relu());
#if 1
for (int hs : { 128, 64, 80, }) {
for (int hs : { 128, 256, 64, 80, }) {
for (int nh : { 32, }) {
for (int kv : { 512, 1024, 2048, 4096, }) {
for (int nb : { 1, 2, 4, 8, 512, 1024, 2048, }) {