CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-04 09:32:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

[wip] chat now has parameter and cfg

Browse Source
This commit is contained in:
Henri Vasserman
2023-07-20 15:37:31 +03:00
parent 082dd81286
commit e4db70720d
2 changed files with 79 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/server/server.cpp
View File

@@ -398,7 +398,7 @@ struct llama_server_context {
evaluator.evaluate(params.n_threads, params.n_batch);
if (cfg_enabled) {
evaluator_guidance.evaluate(params.n_threads, params.n_batch);
evaluator_guidance.evaluate(params.n_threads, params.n_batch);
}
if (params.n_predict == 0) {
@@ -1067,7 +1067,7 @@ int main(int argc, char ** argv) {
llama.loadPrompt();
llama.beginCompletion();
if (llama.params.cfg_negative_prompt.size() > 0) {
if (llama.params.cfg_scale > 1.0f && llama.params.cfg_negative_prompt.size() > 0) {
llama.cfg_enabled = true;
llama.loadGuidancePrompt();
}