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

cuda : increase C to 128 for better performance

Browse source
This commit is contained in:
Georgi Gerganov 2024-02-01 16:12:56 +02:00
parent 9a5c2a1681
commit ac26f27028
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
4 changed files with 37 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml.c
View file

@ -5089,7 +5089,7 @@ static struct ggml_tensor * ggml_soft_max_impl(
GGML_ASSERT(ggml_is_contiguous(mask));
GGML_ASSERT(mask->ne[2] == 1);
GGML_ASSERT(mask->ne[3] == 1);
GGML_ASSERT(ggml_can_repeat_rows(mask, a));
GGML_ASSERT(mask->ne[1] >= a->ne[1]);
}
bool is_node = false;