Skip to content

Fix GLM 4.7 Lite MoE gating func #18980

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
pwilkin merged 3 commits into from
Jan 21, 2026
Merged

Fix GLM 4.7 Lite MoE gating func #18980

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0eab651
Fix GLM 4.7 MoE gating func
pwilkin Jan 21, 2026
2bd7ae9
Update src/models/deepseek2.cpp
pwilkin Jan 21, 2026
a4c9b36
Update src/llama-model.cpp
pwilkin Jan 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion src/llama-model.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1713,7 +1713,11 @@ void llama_model::load_hparams(llama_model_loader & ml) {
if (hparams.expert_gating_func == LLAMA_EXPERT_GATING_FUNC_TYPE_NONE) {
// for compatibility with existing DeepSeek V2 and V2.5 GGUFs
// that have no expert_gating_func model parameter set
hparams.expert_gating_func = LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX;
if (hparams.n_layer == 47 || hparams.n_layer == 48) { // GLM 4.7 Lite
Comment thread
pwilkin marked this conversation as resolved.
Outdated
hparams.expert_gating_func = LLAMA_EXPERT_GATING_FUNC_TYPE_SIGMOID;
} else {
hparams.expert_gating_func = LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX;
}
}

if (ml.get_key(LLM_KV_ROPE_SCALING_YARN_LOG_MUL, hparams.rope_yarn_log_mul, 0.0f)) {
Expand Down
2 changes: 1 addition & 1 deletion src/models/deepseek2.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,7 +215,7 @@ llm_build_deepseek2::llm_build_deepseek2(const llama_model & model, const llm_gr
model.layers[il].ffn_exp_probs_b,
n_expert, n_expert_used,
LLM_FFN_SILU, hparams.expert_weights_norm,
hparams.expert_weights_scale, hparams.expert_weights_scale,
true, hparams.expert_weights_scale,
Comment thread
pwilkin marked this conversation as resolved.
Outdated
(llama_expert_gating_func_type) hparams.expert_gating_func,
il);
cb(moe_out, "ffn_moe_out", il);
Expand Down
Loading