vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

examples : replace llama_kv_cache_seq_* with llama_past_seq_*

Browse source
This commit is contained in:
Francis Couture-Harpin 2024-06-10 14:44:42 -04:00
parent 372482dffe
commit 43d8d4bf9e
23 changed files with 125 additions and 112 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/batched/batched.cpp
View file

@ -112,7 +112,7 @@ int main(int argc, char ** argv) {
// assign the system KV cache to all parallel sequences
// this way, the parallel sequences will "reuse" the prompt tokens without having to copy them
for (int32_t i = 1; i < n_parallel; ++i) {
llama_kv_cache_seq_cp(ctx, 0, i, -1, -1);
llama_past_seq_cp(ctx, 0, i, -1, -1);
}
if (n_parallel > 1) {