cturan/llama.cpp @ cb1c0727bd59803b439b6a3af121c99e6393ff3d

огледало от https://github.com/cturan/llama.cpp

Rand Xie 65cdf34bdc llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)		преди 2 години
..
CMakeLists.txt	b1f4290953 cmake : install targets (#2256)	преди 2 години
save-load-state.cpp	65cdf34bdc llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)	преди 2 години