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

llama : add support for Qwen3 MoE tied word embeddings (#13768)

Browse source
This commit is contained in:
Piotr Jasiukajtis 2025-05-25 10:29:43 +02:00 • committed by GitHub
parent 515fdbf7ed
commit 4032ca4066
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 5 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
src/llama-model.cpp
View file

@ -2489,7 +2489,11 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
// output // output
output_norm = create_tensor(tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd}, 0); output_norm = create_tensor(tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd}, 0);
output = create_tensor(tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}, 0); output = create_tensor(tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}, TENSOR_NOT_REQUIRED);
// if output is NULL, init from the input tok embed
if (output == NULL) {
output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED);
}
for (int i = 0; i < n_layer; ++i) { for (int i = 0; i < n_layer; ++i) {
auto & layer = layers[i]; auto & layer = layers[i];