Inicio Explorar Axuda
Iniciar sesión
cturan
/
llama.cpp
réplica de https://github.com/cturan/llama.cpp
1
0
Fork 0
Ficheiros Incidencias 0 Wiki
Árbore: 1e920018d3
Ramas Etiquetas
llama.cpp
/
examples
/
lookup
Clint Herron 07a3fc0608 Removes multiple newlines at the end of files that is breaking the editorconfig step of CI. (#8258) hai 1 ano
..
CMakeLists.txt 1c641e6aac `build`: rename main → llama-cli, server → llama-server, llava-cli → llama-llava-cli, etc... (#7809) hai 1 ano
README.md 07a3fc0608 Removes multiple newlines at the end of files that is breaking the editorconfig step of CI. (#8258) hai 1 ano
lookup-create.cpp 1442677f92 common : refactor cli arg parsing (#7675) hai 1 ano
lookup-merge.cpp 1c641e6aac `build`: rename main → llama-cli, server → llama-server, llava-cli → llama-llava-cli, etc... (#7809) hai 1 ano
lookup-stats.cpp 1442677f92 common : refactor cli arg parsing (#7675) hai 1 ano
lookup.cpp 1442677f92 common : refactor cli arg parsing (#7675) hai 1 ano

README.md

llama.cpp/examples/lookup

Demonstration of Prompt Lookup Decoding

https://github.com/apoorvumang/prompt-lookup-decoding

The key parameters for lookup decoding are ngram_min, ngram_max and n_draft. The first two determine the size of the ngrams to search for in the prompt for a match. The latter specifies how many subsequent tokens to draft if a match is found.

More info:

https://github.com/ggerganov/llama.cpp/pull/4484 https://github.com/ggerganov/llama.cpp/issues/4226