ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

speculative : update default params (#11954)

Browse source 
* speculative : update default params

* speculative : do not discard the last drafted token
This commit is contained in:
Georgi Gerganov 2025-02-19 13:29:42 +02:00 • committed by GitHub
parent 9626d9351a
commit abd4d0bc4f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 9 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/server/server.cpp
View file

@ -274,7 +274,7 @@ struct server_task {
params.speculative.p_min = json_value(data, "speculative.p_min", defaults.speculative.p_min);
params.speculative.n_min = std::min(params.speculative.n_max, params.speculative.n_min);
params.speculative.n_min = std::max(params.speculative.n_min, 2);
params.speculative.n_min = std::max(params.speculative.n_min, 0);
params.speculative.n_max = std::max(params.speculative.n_max, 0);
// Use OpenAI API logprobs only if n_probs wasn't provided