]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 5dcdf94)
llama : cache llama_token_to_piece (#7587)
authorGeorgi Gerganov <redacted>
Thu, 30 May 2024 16:01:41 +0000 (19:01 +0300)
committerGitHub <redacted>
Thu, 30 May 2024 16:01:41 +0000 (02:01 +1000)
commit5921b8f089d3b7bda86aac5a66825df6a6c10603
tree7ecf18258dd34a32b599c70a11108541311ba3d2tree
parent5dcdf946764fae49a8e2a90bf2f0960bde1c44e8commit | diff
llama : cache llama_token_to_piece (#7587)

* llama : cache llama_token_to_piece

ggml-ci

* llama : use vectors and avoid has_cache

ggml-ci

* llama : throw on unknown tokenizer types

ggml-ci

* llama : print a log of the total cache size
llama.cpp diff | blob | history
llama.h diff | blob | history
Packaging of ggml-org/llama.cpp