Skip to content

glm5next: avoid soft_max gridDim.y overflow in the indexer #214

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

Open
AIalliAI wants to merge 2 commits into
base: glm5next/upstream
Choose a base branch
from
+3 −1
Open

glm5next: avoid soft_max gridDim.y overflow in the indexer #214

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2dc7307
glm5next: avoid soft_max gridDim.y overflow in the indexer
AIalliAI Sep 11, 2026
d798de9
glm5next: clarify soft_max gridDim comment
AIalliAI Sep 11, 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
4 changes: 3 additions & 1 deletion src/models/glm5next.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -383,7 +383,9 @@ ggml_tensor * llama_model_glm5next::graph::build_indexer(
ggml_tensor * ape = ggml_cont(ctx0, ggml_transpose(ctx0, layer.indexer_comp_ape));
gate_t = ggml_add(ctx0, gate_t, ggml_reshape_4d(ctx0, ape, r, d_idx, 1, 1));

ggml_tensor * probs = ggml_soft_max(ctx0, gate_t);
// soft_max maps ne2/ne3 to gridDim.y/z (65535 cap); fold into ne1
ggml_tensor * probs = ggml_soft_max(ctx0, ggml_reshape_2d(ctx0, gate_t, r, d_idx*n_new_max*n_stream));
probs = ggml_reshape_4d(ctx0, probs, r, d_idx, n_new_max, n_stream);
cb(probs, "indexer_pool_probs", il);

ggml_tensor * pool_new = ggml_sum_rows(ctx0, ggml_mul(ctx0, keys_t, probs));
Expand Down