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

auto scale

Browse source
This commit is contained in:
ngxson 2024-07-15 11:41:18 +02:00
parent 703573f608
commit 42415a4874
4 changed files with 36 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

5
common/common.cpp
View file

@ -684,7 +684,7 @@ bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_pa
}
if (arg == "--lora") {
CHECK_ARG
params.lora_adapter.emplace_back(argv[i], 1.0f);
params.lora_adapter.emplace_back(argv[i], 0.0f);
return true;
}
if (arg == "--lora-scaled") {
@ -2089,6 +2089,9 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
llama_free_model(model);
return std::make_tuple(nullptr, nullptr);
}
if (lora_scale == 0.0f) {
lora_scale = llama_lora_adapter_get_default_scale(adapter);
}
llama_lora_adapter_set(lctx, adapter, lora_scale);
}