]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
vocab : refactor tokenizer to reduce init overhead (#9449)
authorZhenwei Jin <redacted>
Sat, 28 Sep 2024 12:10:58 +0000 (20:10 +0800)
committerGitHub <redacted>
Sat, 28 Sep 2024 12:10:58 +0000 (15:10 +0300)
commit6102037bbb55521880ae78a6ee6c2a0c00c901df
treea6936ee87d99f557fda32e21d6b7e7cb19426194
parent9a913110cf471a8287ac06c43cbe307d3cf6df99
vocab : refactor tokenizer to reduce init overhead (#9449)

* refactor tokenizer

* llama : make llm_tokenizer more private

ggml-ci

* refactor tokenizer

* refactor tokenizer

* llama : make llm_tokenizer more private

ggml-ci

* remove unused files

* remove unused fileds to avoid unused filed build error

* avoid symbol link error

* Update src/llama.cpp

* Update src/llama.cpp

---------

Co-authored-by: Georgi Gerganov <redacted>
examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp
src/llama-vocab.cpp
src/llama-vocab.h
src/llama.cpp
tests/test-tokenizer-0.cpp