root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-13 20:14:29 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

llama : use const ref for print_f and fix division by zero

Browse Source
This commit is contained in:
Francis Couture-Harpin 2024-07-04 11:39:32 -04:00
parent 199d0fb0c9
commit 269e07bb00
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/llama.cpp
View File

@ -5688,7 +5688,7 @@ static void llm_load_print_meta(llama_model_loader & ml, llama_model & model) {
const char * rope_scaling_type = LLAMA_ROPE_SCALING_TYPES.at(hparams.rope_scaling_type_train);
auto print_f = [](std::function<uint32_t(uint32_t)> f, uint32_t n) {
auto print_f = [](const std::function<uint32_t(uint32_t)> & f, uint32_t n) {
bool is_var = false;
std::vector<uint32_t> v;
@ -5954,7 +5954,7 @@ static bool llm_load_tensors(
// create tensors for the weights
{
const int64_t n_embd = hparams.n_embd;
const int64_t n_embd_head = n_embd / hparams.n_head();
const int64_t n_embd_head = hparams.n_head() > 0 ? n_embd / hparams.n_head() : 0;
const int64_t n_embd_k_gqa = hparams.n_embd_k_gqa();
const int64_t n_embd_v_gqa = hparams.n_embd_v_gqa();
const int64_t n_embd_gqa = n_embd_v_gqa;