Skip to content

Implement QK attention head chunking for CSA [Deepseek v4] - #5116

Open
octatrifan wants to merge 3 commits into
octatrifan-dsv4-indexer-lossfrom
octatrifan-dsv4-qk-chunking
Open

Implement QK attention head chunking for CSA [Deepseek v4]#5116
octatrifan wants to merge 3 commits into
octatrifan-dsv4-indexer-lossfrom
octatrifan-dsv4-qk-chunking

Address review nits in deepseek_v4_indexer_loss_test.py

ad65b6c
Select commit
Loading
Failed to load commit list.
Google CLA / cla/google succeeded Sep 4, 2026 in 16s

✅ All contributors are covered under a CLA with Google

See https://cla.developers.google.com/ for more info about Google's Contributor License Agreement (CLA).

ℹ️ Googlers: Go here to view more details and manage scans for this pull request.

Details

The following contributors were found for this pull request:

ad65b6c Author: @octatrifan <oct*****an​@gmail.com>, <oct*****an​@google.com>

(Only the first commit for a unique contributor is listed.)