Owner
wdlctc
3 indexed repositories · View on GitHub
-
open-attention-residuals
Open implementation of Attention Residuals (Kimi Team, arXiv:2603.15031)
Python · 83 stars
-
delta-attention-residuals-code
Delta Attention Residuals - supplementary code and pretrained models
Python · 43 stars
-
multi-head-attention-residuals
Multi-Head Attention Residuals (MHAR): zero-parameter multi-head routing over the depth history — paper + reference implementation
Python · 12 stars
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
-
Difficulty 4/5 3-5 days Newbie friendliness 38/100
-
Difficulty 4/5 3-5 days Newbie friendliness 35/100
wdlctc/delta-attention-residuals-code#1 · 1 comment ·