CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-01 09:01:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : more robust cell_max heuristic + wip shift

Browse Source
This commit is contained in:
Georgi Gerganov
2023-09-18 17:15:25 +03:00
parent 4d76d762ef
commit f015b26689
3 changed files with 39 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/llama-bench/llama-bench.cpp
View File

@@ -977,6 +977,8 @@ int main(int argc, char ** argv) {
test t(inst, lmodel, ctx);
llama_kv_cache_keep_seq(ctx, -1);
// warmup run
if (t.n_prompt > 0) {
test_prompt(ctx, std::min(2, t.n_batch), 0, t.n_batch, t.n_threads);
@@ -986,6 +988,8 @@ int main(int argc, char ** argv) {
}
for (int i = 0; i < params.reps; i++) {
llama_kv_cache_keep_seq(ctx, -1);
uint64_t t_start = get_time_ns();
if (t.n_prompt > 0) {
test_prompt(ctx, t.n_prompt, 0, t.n_batch, t.n_threads);