Skip to content

LLM inference in C/C++

C++ ggml

server : add extra trace log for prompt similarity (#26218)

Georgi Gerganov9a3bf2b84923
diff --git a/tools/server/server-task.cpp b/tools/server/server-task.cpp
index 1fd7cce27b..99e63b05f5 100644
--- a/tools/server/server-task.cpp
+++ b/tools/server/server-task.cpp
@@ -1755,6 +1755,8 @@ bool server_prompt_cache::load(server_prompt & prompt, const server_tokens & tok
const float f_keep_cur = float(lcp_cur) / it->prompt.tokens.size();
const float sim_cur = float(lcp_cur) / tokens_new.size();
+ SRV_TRC(" - prompt with length %7zu, lcp = %7d, f_keep = %.3f, sim = %.3f\n", it->prompt.tokens.size(), lcp_cur, f_keep_cur, sim_cur);
+
// don't trash large prompts
if (f_keep_cur < 0.25f) {
continue;
diff --git a/tools/server/server-task.h b/tools/server/server-task.h
index c3eea2ecb8..411d918079 100644
--- a/tools/server/server-task.h
+++ b/tools/server/server-task.h
@@ -650,7 +650,7 @@ struct server_prompt_cache {
server_prompt_cache_state * alloc(const server_prompt & prompt, size_t state_size_main, size_t state_size_drft);
- bool load(server_prompt & prompt, const server_tokens & tokens_new, llama_context * ctx_main, llama_context * ctx_drft, int32_t id_slot);
+ bool load(server_prompt & prompt, const server_tokens & tokens_new, llama_context * ctx_tgt, llama_context * ctx_dft, int32_t id_slot);
void update();
};