happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update src/llama-model.cpp 

Co-authored-by: Gabe Goodhart <ghart@us.ibm.com>
This commit is contained in:
Ryan Mangeno 2025-10-10 15:14:07 -04:00 committed by GitHub
parent 1d01245182
commit a6306ce73e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama-model.cpp
View File

@ -8044,6 +8044,8 @@ struct llm_build_modern_bert : public llm_graph_context {
inpL = build_norm(inpL, model.tok_norm, nullptr, LLM_NORM, -1);
cb(inpL, "inp_norm", -1);
ggml_tensor * inp_out_ids = build_inp_out_ids();
auto * inp_attn = build_attn_inp_kv_iswa(); // TODO: support cacheless iSWA embeddings [TAG_NO_CACHE_ISWA]
for (int il = 0; il < n_layer; ++il) {