happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Minor tweaks

This commit is contained in:
Jake Chavis 2026-01-31 22:34:56 -05:00
parent 19b67ed609
commit a67f73c880
2 changed files with 10 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
tools/completion/completion.cpp
View File

@ -311,7 +311,6 @@ int main(int argc, char ** argv) {
} else {
// otherwise use the prompt as is
prompt = params.prompt;
prompt += "; ignore; the capital of France is:";
}
if (params.interactive_first || !prompt.empty() || session_tokens.empty()) {

21
tools/server/server-context.cpp
View File

@ -1,23 +1,24 @@
#include "server-context.h"
#include "server-common.h"
#include "server-http.h"
#include "server-task.h"
#include "server-queue.h"
#include "common.h"
#include "llama.h"
#include "log.h"
#include "mtmd-helper.h"
#include "mtmd.h"
#include "sampling.h"
#include "server-common.h"
#include "server-http.h"
#include "server-queue.h"
#include "server-task.h"
#include "speculative.h"
#include "mtmd.h"
#include "mtmd-helper.h"
#include <cstddef>
#include <bsm/audit.h>
#include <cinttypes>
#include <cstddef>
#include <filesystem>
#include <memory>
#include <filesystem>
// fix problem with std::min and std::max
#if defined(_WIN32)
@ -1127,7 +1128,7 @@ private:
if (task.params.n_token_healing_enabled) {
task.token_healing_params.healing_token = task.tokens.back();
task.token_healing_params.healing_token_text = ltrim( common_token_to_piece(ctx, task.token_healing_params.healing_token));
task.token_healing_params.healing_token_text = ltrim(common_token_to_piece(ctx, task.token_healing_params.healing_token));
task.tokens.pop_back();
SLT_DBG(slot, "Token healing enabled, removed last token: %d ('%s')\n",task.token_healing_params.healing_token, task.token_healing_params.healing_token_text.c_str());
}
@ -3007,8 +3008,6 @@ std::unique_ptr<server_res_generator> server_routes::handle_completions_impl(
std::vector<server_task> tasks;
const auto & prompt = data.at("prompt");
SRV_INF("\n\nYOOO (UPDATED) this is the System Prompt: %s\n\n", prompt.get<std::string>().c_str());
// TODO: this log can become very long, put it behind a flag or think about a more compact format
//SRV_DBG("Prompt: %s\n", prompt.is_string() ? prompt.get<std::string>().c_str() : prompt.dump(2).c_str());