llama: Add option to merge gate and exp weights (#19139)

* llama: Add option to merge gate and exp weights * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> * Update convert_hf_to_gguf.py Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com> * update constants.py * add gate_up for the all MoE models * convert: simplify merge tensor condition * update constants.py * reduce number of models, add create_tensor_gate_up helper --------- Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2026-02-26 21:01:08 +08:00
parent ffaafde16f
commit b68d75165a
12 changed files with 135 additions and 44 deletions
@@ -218,7 +218,9 @@ llm_build_deepseek2::llm_build_deepseek2(const llama_model & model, const llm_gr
                LLM_FFN_SILU, hparams.expert_weights_norm,
                hparams.expert_weights_scale, hparams.expert_weights_scale,
                (llama_expert_gating_func_type) hparams.expert_gating_func,
-                il);
+                il,
+                nullptr,
+                model.layers[il].ffn_gate_up_exps);
            cb(moe_out, "ffn_moe_out", il);

            // FFN shared expert