transformers
[Qwen3-next] Fix dimension mismatch in torch_chunk_gated_delta_rule and torch_recurrent_gated_delta_rule (#40963)
#41036

Merged

[Qwen3-next] Fix dimension mismatch in torch_chunk_gated_delta_rule and torch_recurrent_gated_delta_rule (#40963) #41036

vasqu merged 5 commits into huggingface:main from notkisk:fix-mismatched-qwen3-next-dims-in-torch_chunk_gated_delta_rule

vasqu commented on 2025-09-22

notkisk force pushed from 3e2394f2 to ad7b43d2 96 days ago

vasqu approved these changes on 2025-09-22

notkisk force pushed from db58944b to 14bc2524 95 days ago

notkisk requested a review from

vasqu 95 days ago

fix mismatched dims for qwen3 next

198f82da

propagate changes

1962634a

chore: renamed tot_heads to total_sequence_length

7515843b

Apply suggestion from @vasqu

f3ab6fd8

minor fix to modular qwen3 next file

6e89ac06

notkisk force pushed from 9be36c92 to 6e89ac06 94 days ago

vasqu enabled auto-merge (squash) 94 days ago

vasqu merged 80f20e0f into main 94 days ago

Reviewers

vasqu

Assignees

No one assigned

Labels

None yet

Milestone

No milestone