graph : remove redundant scale_w parameter (#20235)
This commit is contained in:
@@ -127,7 +127,6 @@ llm_build_afmoe::llm_build_afmoe(const llama_model & model, const llm_graph_para
|
||||
n_expert, n_expert_used,
|
||||
LLM_FFN_SILU,
|
||||
hparams.expert_weights_norm, // norm_w (route_norm=True)
|
||||
hparams.expert_weights_scale, // scale_w
|
||||
hparams.expert_weights_scale, // w_scale (route_scale=2.826)
|
||||
(llama_expert_gating_func_type) hparams.expert_gating_func,
|
||||
il);
|
||||
|
||||
Reference in New Issue
Block a user