Public/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-11 15:29:03 +02:00
Code Issues Packages Projects Releases Wiki Activity

convert : expand Nemotron H conversion fix (#28689)

Browse Source 
* override function for n_h_l

* narrow change for extracting nested attribute

* simpler change; combines has_moe_params

* Apply suggestion from @CISC

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
This commit is contained in:
Kartik Gulia
2026-09-10 16:11:57 +05:30
committed by GitHub
parent 311d4211bf
commit 8c322d5bc4
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
conversion/nemotron.py
+5 -5
View File
@@ -216,14 +216,14 @@ class NemotronHModel(GraniteHybridModel):
hparams = kwargs.pop("hparams", None)
if hparams is None:
hparams = ModelBase.load_hparams(args[0], self.is_mistral_format)
has_moe_params = (
"num_experts_per_tok" in hparams
or (isinstance(hparams.get("llm_config"), dict) and "num_experts_per_tok" in hparams["llm_config"])
)
llm_config = {**hparams, **(hparams.get("llm_config") or {})}
has_moe_params = "num_experts_per_tok" in llm_config
layers_block_type = llm_config.get("layers_block_type")
if has_moe_params:
self.model_arch = gguf.MODEL_ARCH.NEMOTRON_H_MOE
self.is_moe = True
layers_block_type = hparams.get("layers_block_type")
if layers_block_type is not None:
hparams["num_hidden_layers"] = len(layers_block_type)