Accueil Explorer Aide
Connexion
cturan
/
llama.cpp
miroir de https://github.com/cturan/llama.cpp
1
0
Fork 0
Fichiers Tickets 0 Wiki
Aborescence: 07cd41d096
Branches Tags
llama.cpp
/
examples
/
lookup
Johannes Gäßler 28103f4832 Server: fix seed for multiple slots (#6835) il y a 1 an
..
CMakeLists.txt 50ccaf5eac lookup: complement data from context with general text statistics (#5479) il y a 1 an
README.md 7082d24cec lookup : add prompt lookup decoding example (#4484) il y a 2 ans
lookup-create.cpp 1b67731e18 BERT tokenizer fixes (#6498) il y a 1 an
lookup-merge.cpp 50ccaf5eac lookup: complement data from context with general text statistics (#5479) il y a 1 an
lookup-stats.cpp 28103f4832 Server: fix seed for multiple slots (#6835) il y a 1 an
lookup.cpp 28103f4832 Server: fix seed for multiple slots (#6835) il y a 1 an

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