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

speculative : implement stochastic speculative sampling (#5625)

Browse source 
* (WIP) Implement stochastic speculative decoding

* sample from residual distribution on draft accept failure

* fix #5657: force greedy sampling with probs when temp is 0

* remove p_accept parameter

* fix style

* remove unused variables

* add srand() in speculative.cpp

* replace use of rand() with mt19937 sampling

* fixes based on review (@JohannesGaessler)

* fix r random generation

* randomly select next sequence to verify + fix bug in memory freeing

* fix bug in active_seqs sync

* fix uniform int distribution initialization

* remove warnings from comparison between int and size_t

* check grammar in `llama_sample_probability_distribution_impl`

* remove malloc code by utilizing vectors

* add PR link to README
This commit is contained in:
Minsoo Cheong 2024-03-05 03:24:00 +09:00 • committed by GitHub
parent 4ffcdce2ff
commit 6d341ab6c5
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 260 additions and 61 deletions
Download patch file Download diff file Expand all files Collapse all files

7
common/sampling.h
View file

@ -131,6 +131,13 @@ llama_token llama_sampling_sample(
struct llama_context * ctx_cfg,
int idx = 0);
// returns the probability that token of given id will be sampled
llama_token_data_array llama_sampling_probability_distribution(
struct llama_sampling_context * ctx_sampling,
struct llama_context * ctx_main,
struct llama_context * ctx_cfg,
int idx = 0);
void llama_sampling_accept(
struct llama_sampling_context * ctx_sampling,
struct llama_context * ctx_main,