ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

graph : normalize Q, K, V shapes + sync cross attention (#12449)

Browse source 
* graph : normalize Q, K, V shapes and add comments

ggml-ci

* context : synchronize before getting cross attention data

* model : fix command-r attention norm check
This commit is contained in:
Georgi Gerganov 2025-03-18 21:35:19 +02:00 • committed by GitHub
parent bb115d2bf7
commit 75422e8bc4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 433 additions and 277 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama-context.cpp
View file

@ -1143,6 +1143,8 @@ int llama_context::encode(llama_batch & inp_batch) {
if (model.arch == LLM_ARCH_T5 && t_embd) {
//cross.t_embd = t_embd;
synchronize();
cross.n_embd = t_embd->ne[0];
cross.n_enc = t_embd->ne[1];
cross.v_embd.resize(cross.n_embd*cross.n_enc);