vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : combine repetition, frequency and presence penalties in 1 call

Browse source
This commit is contained in:
Georgi Gerganov 2023-10-20 17:05:46 +03:00
parent cd1e937821
commit 6e6587656f
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
5 changed files with 51 additions and 118 deletions
Download patch file Download diff file Expand all files Collapse all files

10
llama.h
View file

@ -560,19 +560,13 @@ extern "C" {
LLAMA_API void llama_set_rng_seed(struct llama_context * ctx, uint32_t seed);
/// @details Repetition penalty described in CTRL academic paper https://arxiv.org/abs/1909.05858, with negative logit fix.
LLAMA_API void llama_sample_repetition_penalty(
struct llama_context * ctx,
llama_token_data_array * candidates,
const llama_token * last_tokens,
size_t last_tokens_size,
float penalty);
/// @details Frequency and presence penalties described in OpenAI API https://platform.openai.com/docs/api-reference/parameter-details.
LLAMA_API void llama_sample_frequency_and_presence_penalties(
LLAMA_API void llama_sample_repetition_penalties(
struct llama_context * ctx,
llama_token_data_array * candidates,
const llama_token * last_tokens,
size_t last_tokens_size,
float repeat_penalty,
float alpha_frequency,
float alpha_presence);