llama.cpp

Fix missing parameters in `llama_init_from_gpt_params`

#1293

Merged