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

llama.cpp: Include the changes from #6122 to exclude the unused outputs of the last layers.

Browse source
This commit is contained in:
root 2024-03-27 04:22:09 +00:00
parent 3c0b830808
commit e4a16f2493
1 changed files with 7 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

7
llama.cpp
View file

@ -6525,6 +6525,13 @@ struct llm_build_context {
cb(cur, "kqv_out", il);
}
if (il == n_layer - 1) {
// skip computing output for unused tokens
struct ggml_tensor * inp_out_ids = build_inp_out_ids();
cur = ggml_get_rows(ctx0, cur, inp_out_ids);
inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);
}
struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA);
cb(ffn_inp, "ffn_inp", il);