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

Merge f4f5b7ac56 into c81f3bbb05

Browse source
This commit is contained in:
Maximilian Winter 2024-10-10 00:07:42 +01:00 • committed by GitHub
commit 3de720c630
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/server/server.cpp
View file

@ -2102,7 +2102,7 @@ struct server_context {
// reuse any previously computed tokens that are common with the new prompt // reuse any previously computed tokens that are common with the new prompt
slot.n_past = common_part(slot.cache_tokens, prompt_tokens); slot.n_past = common_part(slot.cache_tokens, prompt_tokens);
// push the prompt into the sampling context (do not apply grammar) // push the prompt into the sampling context (do not apply grammar)
for (int i = 0; i < slot.n_past; ++i) { for (int i = 0; i < slot.n_past; ++i) {
gpt_sampler_accept(slot.smpl, slot.cache_tokens[i], false); gpt_sampler_accept(slot.smpl, slot.cache_tokens[i], false);