diff options
| author | saood06 <saood05@gmail.com> | 2025-01-23 10:24:10 -0600 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-01-23 18:24:10 +0200 |
| commit | 2195632581c4f52707059b5963fe622ccead0dd2 (patch) | |
| tree | 34d46a344c5d32ff699126cea9255eb13fd3b38a /gguf-py/gguf/gguf_writer.py | |
| parent | c2624b2fd324ff98cc137397f5b0e1d22869cb58 (diff) | |
Deepseek V3 support added (#176)
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Diffstat (limited to 'gguf-py/gguf/gguf_writer.py')
| -rw-r--r-- | gguf-py/gguf/gguf_writer.py | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/gguf-py/gguf/gguf_writer.py b/gguf-py/gguf/gguf_writer.py index 76385a82..e31bf97b 100644 --- a/gguf-py/gguf/gguf_writer.py +++ b/gguf-py/gguf/gguf_writer.py @@ -26,6 +26,7 @@ from .constants import ( RopeScalingType, PoolingType, TokenType, + ExpertGatingFuncType, ) from .quants import quant_shape_from_byte_shape @@ -670,6 +671,12 @@ class GGUFWriter: def add_expert_weights_scale(self, value: float) -> None: self.add_float32(Keys.LLM.EXPERT_WEIGHTS_SCALE.format(arch=self.arch), value) + def add_expert_weights_norm(self, value: bool) -> None: + self.add_bool(Keys.LLM.EXPERT_WEIGHTS_NORM.format(arch=self.arch), value) + + def add_expert_gating_func(self, value: ExpertGatingFuncType) -> None: + self.add_uint32(Keys.LLM.EXPERT_GATING_FUNC.format(arch=self.arch), value.value) + def add_layer_norm_eps(self, value: float) -> None: self.add_float32(Keys.Attention.LAYERNORM_EPS.format(arch=self.arch), value) |