Canales actualizados recientemente

Elige uno para ver solo ese canalVer todos los canales →
Matthew Bermanhoy +242AI Engineerhoy +6The Buildershoy +120VC with Harry Stebbingsayer +351Jane Streetayer +119Y Combinatorayer +1硅谷101anteayerPragerUhace 3 díasa16zhace 4 días +1SemiAnalysishace 4 días

Resultados

1 artículos
Ordenar
CanalJane StreetTemaTransformers & attention
7 diapositivas1:16:29EN
Jane Street· 26 ago 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 & attentionMathematicsFundamentos de ML+1