RSS

eggsyntax

Karma: 2,514

AI safety & alignment researcher

In Rob Bensinger’s typology: AGI-alarmed, tentative welfarist, and eventualist (or variabilist for sufficiently long values of ‘soon’).

I have signed no contracts or agreements whose existence I cannot mention.

Your LLM-as­sisted sci­en­tific break­through prob­a­bly isn’t real

eggsyntax2 Sep 2025 15:05 UTC
143 points
39 comments7 min readLW link

On the func­tional self of LLMs

eggsyntax7 Jul 2025 15:39 UTC
105 points
35 comments8 min readLW link

Show, not tell: GPT-4o is more opinionated in images than in text

2 Apr 2025 8:51 UTC
112 points
41 comments3 min readLW link

Num­ber­wang: LLMs Do­ing Au­tonomous Re­search, and a Call for Input

16 Jan 2025 17:20 UTC
71 points
30 comments31 min readLW link

LLMs Look In­creas­ingly Like Gen­eral Reasoners

eggsyntax8 Nov 2024 23:47 UTC
94 points
45 comments3 min readLW link

AIS ter­minol­ogy pro­posal: stan­dard­ize terms for prob­a­bil­ity ranges

eggsyntax30 Aug 2024 15:43 UTC
30 points
12 comments2 min readLW link

LLM Gen­er­al­ity is a Timeline Crux

eggsyntax24 Jun 2024 12:52 UTC
219 points
119 comments7 min readLW link

Lan­guage Models Model Us

eggsyntax17 May 2024 21:00 UTC
159 points
55 comments7 min readLW link

Use­ful start­ing code for interpretability

eggsyntax13 Feb 2024 23:13 UTC
26 points
2 comments1 min readLW link

eggsyn­tax’s Shortform

eggsyntax13 Jan 2024 22:34 UTC
3 points
239 comments1 min readLW link