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 : extend llama_kv_cache API

Browse Source
This commit is contained in:
Georgi Gerganov
2023-09-18 15:53:03 +03:00
parent 6952a460b9
commit 4d76d762ef
4 changed files with 84 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/embd-input/embd-input-lib.cpp
View File

@@ -79,7 +79,7 @@ bool eval_float(void * model, float * input, int N){
if (n_eval > n_batch) {
n_eval = n_batch;
}
llama_batch batch = { uint32_t(n_eval), nullptr, (input+i*n_emb), nullptr, nullptr, n_past, 1, 0, false };
llama_batch batch = { uint32_t(n_eval), nullptr, (input+i*n_emb), nullptr, nullptr, n_past, 1, 0, };
if (llama_decode(ctx, batch, params.n_threads)) {
fprintf(stderr, "%s : failed to eval\n", __func__);
return false;