LESSWRONG
LW

370

Apollo Research (org)

This page is a stub.

1

1

Posts tagged Apollo Research (org)

229SAE feature geometry is outside the superposition hypothesis

1y

17

217Announcing Apollo Research

Marius Hobbhahn, beren, Lee Sharkey, Lucius Bushnaq, Dan Braun, Mikita Balesni, Jérémy Scheurer

2y

11

166You can remove GPT2’s LayerNorm by fine-tuning for an hour

1y

11

123A List of 45+ Mech Interp Project Ideas from Apollo Research’s Interpretability Team

Lee Sharkey, Lucius Bushnaq, Dan Braun, StefanHex, Nicholas Goldowsky-Dill

1y

18

109Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs

L Rudolf L, bilalchughtai, Jan Betley, kaivu, Jérémy Scheurer, Mikita Balesni, AlexMeinke, Owain_Evans, Marius Hobbhahn

1y

37

108Attribution-based parameter decomposition

Lucius Bushnaq, Dan Braun, StefanHex, jake_mendel, Lee Sharkey

8mo

22

108The 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, Marius Hobbhahn

1y

4

96Sparsify: A mechanistic interpretability research agenda

1y

23

93Apollo Research 1-year update

Marius Hobbhahn, Lee Sharkey, Lucius Bushnaq, Dan Braun, Mikita Balesni, Jérémy Scheurer, Nicholas Goldowsky-Dill, StefanHex, jake_mendel, AlexMeinke, rusheb

1y

0

72We need a Science of Evals

Marius Hobbhahn, Jérémy Scheurer

2y

13

65[Interim research report] Activation plateaus & sensitive directions in GPT2

StefanHex, jake_mendel

1y

2

65An Opinionated Evals Reading List

Marius Hobbhahn, Jérémy Scheurer

1y

0

64Understanding strategic deception and deceptive alignment

Marius Hobbhahn, Mikita Balesni, Jérémy Scheurer, Dan Braun

2y

16

57Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning

Dan Braun, Jordan Taylor, Nicholas Goldowsky-Dill, Lee Sharkey

1y

20

55A starter guide for evals

Marius Hobbhahn, Jérémy Scheurer, Mikita Balesni, rusheb, AlexMeinke

2y

2

Load More (15/22)

Add Posts