]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
llama : improve infill support and special token detection (#9798)
authorGeorgi Gerganov <redacted>
Sat, 12 Oct 2024 05:21:51 +0000 (08:21 +0300)
committerGitHub <redacted>
Sat, 12 Oct 2024 05:21:51 +0000 (08:21 +0300)
commit11ac9800aff532715a5bc7991062c68ba3472e6e
tree526fe6d94d82c126f4497cbea986df8fcb910eb4
parent943d20b4111c746bcd9dbc7e4771de313b08b50c
llama : improve infill support and special token detection (#9798)

* llama : improve infill support

ggml-ci

* llama : add more FIM token strings

ggml-ci

* server : update prompt on slot restore (#9800)

* gguf : deprecate old FIM token KVs
12 files changed:
common/arg.cpp
common/common.cpp
common/common.h
examples/infill/infill.cpp
examples/server/README.md
examples/server/server.cpp
gguf-py/gguf/constants.py
gguf-py/gguf/gguf_writer.py
include/llama.h
src/llama-vocab.cpp
src/llama-vocab.h
src/llama.cpp