RSS

jake_mendel

Karma: 344

Interpretability Researcher at Apollo Research

In­ter­pretabil­ity: In­te­grated Gra­di­ents is a de­cent at­tri­bu­tion method

Lucius Bushnaq, jake_mendel, StefanHex and Kaarel
20 May 2024 17:55 UTC
14 points
7 comments6 min readLW link

The Lo­cal In­ter­ac­tion Ba­sis: Iden­ti­fy­ing Com­pu­ta­tion­ally-Rele­vant and Sparsely In­ter­act­ing Fea­tures in Neu­ral Networks

Lucius Bushnaq, jake_mendel, Dan Braun, StefanHex, Nicholas Goldowsky-Dill, Kaarel, Avery, Joern Stoehler, debrevitatevitae, Magdalena Wache and Marius Hobbhahn
20 May 2024 17:53 UTC
97 points
2 comments3 min readLW link

A start­ing point for mak­ing sense of task struc­ture (in ma­chine learn­ing)

Kaarel, RP and jake_mendel
24 Feb 2024 1:51 UTC
39 points
2 comments12 min readLW link

Toward A Math­e­mat­i­cal Frame­work for Com­pu­ta­tion in Superposition

Dmitry Vaintrob, jake_mendel and Kaarel
18 Jan 2024 21:06 UTC
184 points
17 comments73 min readLW link