]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: dacfceb)
ci : use Pythia models instead of OpenLlama (#7470)
authorGeorgi Gerganov <redacted>
Thu, 23 May 2024 12:28:14 +0000 (15:28 +0300)
committerGitHub <redacted>
Thu, 23 May 2024 12:28:14 +0000 (15:28 +0300)
commit55ac3b7aeaf52f19786ed96e885d89521fc0f6c8
tree83f235f4e9c19ce6a46e7ef621799f2454239ca4tree
parentdacfcebd6022175848e978f82811a244f1033038commit | diff
ci : use Pythia models instead of OpenLlama (#7470)

* ci : start using Pythia models over OpenLlama

ggml-ci

* ci : disable q2_k ppl tests

* ci : use convert-hf-to-gguf.py

* ci : update gg_get_model

* ci : fix convert outfile name

ggml-ci

* llama : gptneox arch use F32 attn prec

ggml-ci
ci/run.sh diff | blob | history
llama.cpp diff | blob | history
Packaging of ggml-org/llama.cpp