diff options
Diffstat (limited to 'llama.cpp/examples/lookup/README.md')
| -rw-r--r-- | llama.cpp/examples/lookup/README.md | 12 |
1 files changed, 12 insertions, 0 deletions
diff --git a/llama.cpp/examples/lookup/README.md b/llama.cpp/examples/lookup/README.md new file mode 100644 index 0000000..07d7384 --- /dev/null +++ b/llama.cpp/examples/lookup/README.md @@ -0,0 +1,12 @@ +# 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/ggml-org/llama.cpp/pull/4484 +https://github.com/ggml-org/llama.cpp/issues/4226 |