cturan/llama.cpp @ 34a14b28ff7f3c98730339bacee035091b2a812a

réplica de https://github.com/cturan/llama.cpp

Rand Xie 65cdf34bdc llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)		%!s(int64=2) %!d(string=hai) anos
..
CMakeLists.txt	b1f4290953 cmake : install targets (#2256)	%!s(int64=2) %!d(string=hai) anos
save-load-state.cpp	65cdf34bdc llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)	%!s(int64=2) %!d(string=hai) anos