ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-01-26 17:20:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

All the others

Browse Source
This commit is contained in:
Iwan Kawrakow
2026-01-12 16:22:53 +00:00
committed by Kawrakow
parent c771666d04
commit 0a18f1fadd
1 changed files with 8 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
src/llama-build-context.cpp
View File

@@ -1941,7 +1941,7 @@ ggml_cgraph * llm_build_context::build_llama() {
LLM_FFN_SILU, false,
false, 0.0,
LLM_EXPERT_GATING_FUNC_SIGMOID,
cb, il, gf, true);
cb, il, gf, true, model.layers[il].ffn_up_gate_exps);
// Shared experts
ggml_tensor * shexp_out = llm_build_ffn(ctx0, lctx, nullptr, ffn_inp_normed,
@@ -2774,7 +2774,7 @@ ggml_cgraph * llm_build_context::build_dbrx() {
LLM_FFN_SILU, true,
false, 0.0,
LLM_EXPERT_GATING_FUNC_SOFTMAX,
cb, il, gf);
cb, il, gf, false, model.layers[il].ffn_up_gate_exps);
cb(cur, "ffn_moe_out", il);
cur = ggml_add(ctx0, cur, ffn_inp);
@@ -3862,7 +3862,7 @@ ggml_cgraph * llm_build_context::build_qwen2moe() {
LLM_FFN_SILU, false,
false, 0.0,
LLM_EXPERT_GATING_FUNC_SOFTMAX,
cb, il, gf);
cb, il, gf, false, model.layers[il].ffn_up_gate_exps);
cb(cur, "ffn_moe_out", il);
// FFN shared expert
@@ -6784,7 +6784,7 @@ ggml_cgraph * llm_build_context::build_deepseek2() {
LLM_FFN_SILU, hparams.expert_weights_norm,
true, hparams.expert_weights_scale,
(enum llm_expert_gating_func_type) hparams.expert_gating_func,
cb, il, gf);
cb, il, gf, false, model.layers[il].ffn_up_gate_exps);
cb(moe_out, "ffn_moe_out", il);
// FFN shared expert
@@ -8017,7 +8017,7 @@ ggml_cgraph * llm_build_context::build_dots1() {
LLM_FFN_SILU, hparams.expert_weights_norm,
true, hparams.expert_weights_scale,
(enum llm_expert_gating_func_type) hparams.expert_gating_func,
cb, il, gf);
cb, il, gf, false, model.layers[il].ffn_up_gate_exps);
cb(moe_out, "ffn_moe_out", il);
{
@@ -8287,7 +8287,7 @@ ggml_cgraph * llm_build_context::build_hunyuan_moe() {
n_expert, n_expert_used,
LLM_FFN_SILU, true, false, 0.0f,
LLM_EXPERT_GATING_FUNC_SOFTMAX,
LLM_FFN_SILU, cb, il, gf, true);
LLM_FFN_SILU, cb, il, gf, true, model.layers[il].ffn_up_gate_exps);
cur = lctx.cvec.apply_to(ctx0, cur, il);
cb(cur, "l_out", il);
@@ -8360,7 +8360,7 @@ ggml_cgraph * llm_build_context::build_mimo2() {
n_expert, n_expert_used,
LLM_FFN_SILU, true, false, 0.0f,
LLM_EXPERT_GATING_FUNC_SIGMOID,
LLM_FFN_SILU, cb, il, gf, true);
LLM_FFN_SILU, cb, il, gf, true, model.layers[il].ffn_up_gate_exps);
}
cur = lctx.cvec.apply_to(ctx0, cur, il);
@@ -8537,7 +8537,7 @@ ggml_cgraph * llm_build_context::build_bailingmoe2() {
LLM_FFN_SILU, hparams.expert_weights_norm,
true, hparams.expert_weights_scale,
(llm_expert_gating_func_type) hparams.expert_gating_func,
cb, il, gf);
cb, il, gf, false, model.layers[il].ffn_up_gate_exps);
cb(moe_out, "ffn_moe_out", il);
ggml_tensor * ffn_shexp = llm_build_ffn(ctx0, lctx, nullptr, cur,