]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)
authorRand Xie <redacted>
Fri, 28 Jul 2023 08:42:53 +0000 (01:42 -0700)
committerGitHub <redacted>
Fri, 28 Jul 2023 08:42:53 +0000 (11:42 +0300)
commit65cdf34bdc469fa86248e667a5880992684ef114
treef58f467c3f33f93819d2d45e8fbdf7cf630a45ec
parentedcc7ae7d26007bbf83136e9d33f863fcad9b871
llama : use n_embd_gqa instead of n_embd to handle llama-2 70B (#2433)
examples/save-load-state/save-load-state.cpp
llama.cpp