mirror of
https://github.com/ggml-org/whisper.cpp.git
synced 2026-10-06 14:31:29 +02:00
talk-llama : sync llama.cpp
This commit is contained in:
@@ -131,7 +131,7 @@ llama_model_command_r::graph::graph(const llama_model & model, const llm_graph_p
|
||||
res->t_embd = cur;
|
||||
|
||||
// lm_head
|
||||
cur = build_lora_mm(model.output, cur);
|
||||
cur = build_lora_mm(model.output, cur, model.output_s);
|
||||
|
||||
if (f_logit_scale) {
|
||||
cur = ggml_scale(ctx0, cur, f_logit_scale);
|
||||
|
||||
Reference in New Issue
Block a user