Skip to content

convert : correct gemma 3n conversion #14450

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 3, 2025
Merged

convert : correct gemma 3n conversion #14450

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b7f5024
convert : correct gemma 3n conversion
ngxson Jun 29, 2025
6692656
rm redundant code
ngxson Jun 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions gguf-py/gguf/gguf_writer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -714,8 +714,8 @@ def add_max_alibi_bias(self, bias: float) -> None:
def add_clamp_kqv(self, value: float) -> None:
self.add_float32(Keys.Attention.CLAMP_KQV.format(arch=self.arch), value)

def add_shared_kv_layers(self, value: float) -> None:
self.add_float32(Keys.Attention.SHARED_KV_LAYERS.format(arch=self.arch), value)
def add_shared_kv_layers(self, value: int) -> None:
self.add_uint32(Keys.Attention.SHARED_KV_LAYERS.format(arch=self.arch), value)

def add_sliding_window_pattern(self, value: Sequence[bool]) -> None:
self.add_array(Keys.Attention.SLIDING_WINDOW_PATTERN.format(arch=self.arch), value)
Expand Down
3 changes: 3 additions & 0 deletions src/llama-quant.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -894,6 +894,9 @@ static void llama_model_quantize_impl(const std::string & fname_inp, const std::
if (params->token_embedding_type < GGML_TYPE_COUNT && strcmp(tensor->name, "token_embd.weight") == 0) {
new_type = params->token_embedding_type;
}
if (params->token_embedding_type < GGML_TYPE_COUNT && strcmp(tensor->name, "per_layer_token_embd.weight") == 0) {
new_type = params->token_embedding_type;
}
if (params->output_tensor_type < GGML_TYPE_COUNT && strcmp(tensor->name, "output.weight") == 0) {
new_type = params->output_tensor_type;
}
Expand Down
Loading