sleepy/llama.cpp
1
0
Fork 0
Code Issues 16 Pull Requests Actions 33 Packages Projects Releases Wiki Activity

mtmd: fix use_non_causal being reported incorrectly (#18793)

Browse Source 
* mtmd: fix use_non_causal being reported incorrectly

* move clip_is_mrope to mtmd_decode_use_mrope

* fix sloppy code ggml_cpy
This commit is contained in:
Xuan-Son Nguyen
2026-01-13 12:19:38 +01:00
committed by GitHub
parent 0a57271ab6
commit e047f9ee9d
4 changed files with 15 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/gemma3n-iswa.cpp
+4 -4
View File
@@ -258,12 +258,12 @@ ggml_tensor * llm_build_gemma3n_iswa::get_per_layer_inputs() {
res->add_input(std::move(inp));
} else {
// Vision embedding path: use padding token (ID=0) embedding
// TODO: verify if this is the correct behavior in transformers implementation
const int64_t embd_size = model.tok_embd_per_layer->ne[0]; // n_embd_altup * n_layer
// Extract and dequantize padding token embedding (column 0)
ggml_tensor * padding_q = ggml_view_1d(ctx0, model.tok_embd_per_layer, embd_size, 0);
ggml_tensor * padding_f32 = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, embd_size);
inp_per_layer = ggml_cpy(ctx0, padding_q, padding_f32);
// Extract and dequantize padding token embedding (row 0)
ggml_tensor * padding = ggml_view_1d(ctx0, model.tok_embd_per_layer, embd_size, 0);
inp_per_layer = ggml_cast(ctx0, padding, GGML_TYPE_F32);
// Reshape to [n_embd_altup, n_layer, 1]
inp_per_layer = ggml_reshape_3d(ctx0, inp_per_layer, n_embd_altup, n_layer, 1);