Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

model : load hparams.n_layer_nextn before n_layer() calls (#28159)

Browse Source 
* load hparams.n_layer_nextn before n_layer() calls

* remove duplicate loads
This commit is contained in:
Sigbjørn Skjæret
2026-09-01 13:55:45 +02:00
committed by GitHub
parent fe2120bc9d
commit 9d817213a0
21 changed files with 7 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/models/bailingmoe3.cpp
-1
View File
@@ -22,7 +22,6 @@ void llama_model_bailingmoe3::load_arch_hparams(llama_model_loader & ml) {
ml.get_key(LLM_KV_EXPERT_WEIGHTS_SCALE, hparams.expert_weights_scale, false);
ml.get_key(LLM_KV_EXPERT_WEIGHTS_NORM, hparams.expert_weights_norm, false);
ml.get_key(LLM_KV_EXPERT_GATING_FUNC, hparams.expert_gating_func);
ml.get_key(LLM_KV_NEXTN_PREDICT_LAYERS, hparams.n_layer_nextn, false);
ml.get_key_or_arr(LLM_KV_SWIGLU_CLAMP_EXP, hparams.swiglu_clamp_exp, hparams.n_layer_all, false);
ml.get_key_or_arr(LLM_KV_SWIGLU_CLAMP_SHEXP, hparams.swiglu_clamp_shexp, hparams.n_layer_all, false);