sdgoij/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-05-15 05:24:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

sampling : hide prev behind API and apply #3661

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2023-10-20 18:26:20 +03:00
parent 7e2b5fb1dd
commit 56ba00b923
9 changed files with 119 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/parallel/parallel.cpp
View File

@@ -330,7 +330,7 @@ int main(int argc, char ** argv) {
const llama_token id = llama_sampling_sample(client.ctx_sampling, ctx, NULL, client.i_batch - i);
llama_sampling_accept(client.ctx_sampling, ctx, id);
llama_sampling_accept(client.ctx_sampling, ctx, id, true);
if (client.n_decoded == 1) {
// start measuring generation time after the first token to make sure all concurrent clients