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

llama : add option for greedy sampling with probs

Browse source
This commit is contained in:
Georgi Gerganov 2023-10-27 16:12:01 +03:00
parent 34b2a5e1ee
commit 4aa1fb0d38
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
2 changed files with 7 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/speculative/speculative.cpp
View file

@ -118,7 +118,7 @@ int main(int argc, char ** argv) {
std::vector<seq_draft> drafts(n_seq_dft);
params.sparams.grammar.clear(); // the draft samplers will copy the target sampler's grammar
params.sparams.temp = std::max(0.01f, params.sparams.temp);
params.sparams.temp = 0.0f;
for (int s = 0; s < n_seq_dft; ++s) {
drafts[s].ctx_sampling = llama_sampling_init(params.sparams);