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

llama : gemma3 : use output tensor if it exists in model weight (#12506)

Browse source 
* llama : gemma3 : use output tensor if it exists in model weight

* also add to the llm_tensor_names
This commit is contained in:
Xuan-Son Nguyen 2025-03-22 23:28:19 +01:00 • committed by GitHub
parent ba932dfb50
commit fbdfefe74e
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 8 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

1
gguf-py/gguf/constants.py
View file

@ -1113,6 +1113,7 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
],
MODEL_ARCH.GEMMA3: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_Q_NORM,