Recently updated channels

Pick one to see only that channelBrowse all channels →
Matthew Bermantoday +20AI Engineertoday +12The Builderstoday +1Cognitive Revolution "How AI Changes Everything"yesterday +1Theo - t3․ggyesterday +115Sequoia Capitalyesterday +1a16zyesterday +1Reid Hoffmanyesterday +1Chris Williamsonyesterday +1Conversations with Tyleryesterday

Results

1 articles
Sort
ChannelJane StreetTopicModels & trainingPersonGrant Sanderson
7 slides1:16:29EN
Jane Street· Aug 26, 2026Grant Sanderson and Alok Puranik: Positional Encodings Through Group TheoryGrant Sanderson, Alok Puranik · 3Blue1BrownStarting with assumptions about attention, they derive matrix exponentials and connect rotations, decay and Jordan blocks to RoPE and ALiBi.Transformers & attentionMathematicsML fundamentals+1