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

convert : Add support for Microsoft Phi-4 model (#10817)

Browse source 
* convert : use GPT2 vocab for Phi-4 model

* convert : use null value of sliding_window to distinguish Phi-4 from other PHI3-based models

* llama : do not use sliding window attention mask for Phi-4 model

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
This commit is contained in:
fairydreaming 2024-12-19 10:37:12 +01:00 • committed by GitHub
parent cd920d0ac3
commit 7585edbdeb
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 22 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

10
src/llama.cpp
View file

@ -13333,7 +13333,13 @@ struct llm_build_context {
struct ggml_tensor * inp_pos = build_inp_pos();
// KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask_swa = build_inp_KQ_mask_swa();
struct ggml_tensor * KQ_mask = nullptr;
if (hparams.n_swa == 0) {
// Phi-4 doesn't use sliding window attention
KQ_mask = build_inp_KQ_mask();
} else {
KQ_mask = build_inp_KQ_mask_swa();
}
for (int il = 0; il < n_layer; ++il) {
auto residual = inpL;
@ -13391,7 +13397,7 @@ struct llm_build_context {
cur = llm_build_kv(ctx0, lctx, kv_self, gf,
model.layers[il].wo, model.layers[il].bo,
Kcur, Vcur, Qcur, KQ_mask_swa, n_tokens, kv_head, n_kv, 1.0f, cb, il);
Kcur, Vcur, Qcur, KQ_mask, n_tokens, kv_head, n_kv, 1.0f, cb, il);
}
if (il == n_layer - 1) {