CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-31 08:51:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

speculative : refactor and add a simpler example (#10362)

Browse Source 
* speculative : refactor and add a simpler example

ggml-ci

* speculative : clean-up and add comments and TODOs [no ci]

* speculative : manage context in common_speculative

ggml-ci

* speculative : simplify

ggml-ci

* speculative : simplify (cont)

ggml-ci

* speculative : add --draft-min CLI arg

* speculative : minor fixup

* make : build fixes

* speculative : do not redraft previous drafts

ggml-ci

* speculative : fix the draft sampling

ggml-ci

* speculative : fix compile warning

* common : refactor args

ggml-ci

* common : change defaults [no ci]

* common : final touches

ggml-ci
This commit is contained in:
Georgi Gerganov
2024-11-25 09:58:41 +02:00
committed by GitHub
parent cce5a90075
commit d9d54e498d
28 changed files with 1028 additions and 326 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/lookup/lookup.cpp
View File

@@ -22,7 +22,7 @@ int main(int argc, char ** argv){
common_init();
// max. number of additional tokens to draft if match is found
const int n_draft = params.n_draft;
const int n_draft = params.speculative.n_max;
const bool dump_kv_cache = params.dump_kv_cache;
@@ -102,7 +102,7 @@ int main(int argc, char ** argv){
bool has_eos = false;
struct common_sampler * smpl = common_sampler_init(model, params.sparams);
struct common_sampler * smpl = common_sampler_init(model, params.sampling);
std::vector<llama_token> draft;