RSS

keith_wynroe

Karma: 292

Find­ing an Er­ror-De­tec­tion Fea­ture in Deep­Seek-R1

keith_wynroeApr 24, 2025, 4:03 PM
15 points
0 comments7 min readLW link

De­com­pos­ing the QK cir­cuit with Bilin­ear Sparse Dic­tionary Learning

Jul 2, 2024, 1:17 PM
86 points
7 comments12 min readLW link

An OV-Co­her­ent Toy Model of At­ten­tion Head Superposition

Aug 29, 2023, 7:44 PM
26 points
2 comments6 min readLW link

Liter­a­ture re­view of TAI timelines

Jan 27, 2023, 8:07 PM
35 points
7 comments2 min readLW link
(epochai.org)

You’re Not One “You”—How De­ci­sion The­o­ries Are Talk­ing Past Each Other

keith_wynroeJan 9, 2023, 1:21 AM
28 points
11 comments8 min readLW link