RSS

Con­jec­ture (org)

TagLast edit: 24 Nov 2022 16:27 UTC by Andrea_Miotti

Conjecture is an alignment startup founded by Connor Leahy, Sid Black and Gabriel Alfour, which aims to scale alignment research.

The initial directions of their research agenda include:

We Are Con­jec­ture, A New Align­ment Re­search Startup

Connor Leahy8 Apr 2022 11:40 UTC
186 points
24 comments4 min readLW link

Con­nor Leahy on Dy­ing with Dig­nity, EleutherAI and Conjecture

Michaël Trazzi22 Jul 2022 18:44 UTC
176 points
29 comments14 min readLW link
(theinsideview.ai)

Episte­molog­i­cal Vigilance for Alignment

adamShimi6 Jun 2022 0:27 UTC
52 points
11 comments10 min readLW link
(epistemologicalvigilance.substack.com)

Refine’s First Blog Post Day

adamShimi13 Aug 2022 10:23 UTC
55 points
3 comments1 min readLW link

Simulators

janus2 Sep 2022 12:45 UTC
452 points
102 comments44 min readLW link
(generative.ink)

Con­jec­ture: a ret­ro­spec­tive af­ter 8 months of work

23 Nov 2022 17:10 UTC
179 points
9 comments8 min readLW link

Search­ing for Search

28 Nov 2022 15:31 UTC
55 points
6 comments14 min readLW link

Refine’s Se­cond Blog Post Day

adamShimi20 Aug 2022 13:01 UTC
19 points
0 comments1 min readLW link

What I Learned Run­ning Refine

adamShimi24 Nov 2022 14:49 UTC
103 points
5 comments4 min readLW link

The Sin­gu­lar Value De­com­po­si­tions of Trans­former Weight Ma­tri­ces are Highly Interpretable

28 Nov 2022 12:54 UTC
155 points
25 comments31 min readLW link

No One-Size-Fit-All Epistemic Strategy

adamShimi20 Aug 2022 12:56 UTC
23 points
1 comment2 min readLW link

Shapes of Mind and Plu­ral­ism in Alignment

adamShimi13 Aug 2022 10:01 UTC
30 points
1 comment2 min readLW link

Ab­stract­ing The Hard­ness of Align­ment: Un­bounded Atomic Optimization

adamShimi29 Jul 2022 18:59 UTC
62 points
3 comments16 min readLW link
(epistemologicalvigilance.substack.com)

Levels of Pluralism

adamShimi27 Jul 2022 9:35 UTC
30 points
0 comments14 min readLW link
(epistemologicalvigilance.substack.com)

Ro­bust­ness to Scal­ing Down: More Im­por­tant Than I Thought

adamShimi23 Jul 2022 11:40 UTC
37 points
5 comments3 min readLW link
(epistemologicalvigilance.substack.com)

How to Diver­sify Con­cep­tual Align­ment: the Model Be­hind Refine

adamShimi20 Jul 2022 10:44 UTC
76 points
11 comments8 min readLW link
(epistemologicalvigilance.substack.com)

Mo­saic and Pal­impsests: Two Shapes of Research

adamShimi12 Jul 2022 9:05 UTC
38 points
3 comments9 min readLW link
(epistemologicalvigilance.substack.com)

Refine: An In­cu­ba­tor for Con­cep­tual Align­ment Re­search Bets

adamShimi15 Apr 2022 8:57 UTC
123 points
10 comments4 min readLW link

Cir­cum­vent­ing in­ter­pretabil­ity: How to defeat mind-readers

Lee Sharkey14 Jul 2022 16:59 UTC
92 points
8 comments36 min readLW link

Con­jec­ture: In­ter­nal In­fo­haz­ard Policy

29 Jul 2022 19:07 UTC
118 points
6 comments19 min readLW link

Method­olog­i­cal Ther­apy: An Agenda For Tack­ling Re­search Bottlenecks

22 Sep 2022 18:41 UTC
54 points
6 comments9 min readLW link

Mys­ter­ies of mode collapse

janus8 Nov 2022 10:37 UTC
212 points
32 comments14 min readLW link

Cur­rent themes in mechanis­tic in­ter­pretabil­ity research

16 Nov 2022 14:14 UTC
82 points
3 comments12 min readLW link

Con­jec­ture Se­cond Hiring Round

23 Nov 2022 17:11 UTC
83 points
0 comments1 min readLW link

AMA Con­jec­ture, A New Align­ment Startup

adamShimi9 Apr 2022 9:43 UTC
46 points
41 comments1 min readLW link

Un­der­stand­ing Con­jec­ture: Notes from Con­nor Leahy interview

Akash15 Sep 2022 18:37 UTC
103 points
24 comments15 min readLW link

In­ter­pret­ing Neu­ral Net­works through the Poly­tope Lens

23 Sep 2022 17:58 UTC
122 points
26 comments33 min readLW link

Re-Ex­am­in­ing LayerNorm

Eric Winsor1 Dec 2022 22:20 UTC
85 points
9 comments5 min readLW link

The First Filter

26 Nov 2022 19:37 UTC
52 points
5 comments1 min readLW link

Bi­ases are en­g­ines of cognition

30 Nov 2022 16:47 UTC
40 points
7 comments1 min readLW link