LESSWRONG
LW

Gunnar_Zarncke
10631Ω27141391732
Message
Dialogue
Subscribe

Software engineering, parenting, cognition, meditation, other
Linkedin, Facebook, Admonymous (anonymous feedback)

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
8Gunnar_Zarncke's Shortform
5y
175
[Intuitive self-models] 1. Preliminaries
Gunnar_Zarncke6d140

The sequence has been reviewed by Scott Alexander in Practically-A-Book Review: Byrnes on Trance.

Reply
Raemon's Shortform
Gunnar_Zarncke7d71

I wonder whether this tweet by Yudkowsky is related.

Reply
RohanS's Shortform
Gunnar_Zarncke7d20

Intuitively, when I'm more tired or most stressed. I would guess that is most likely in the morning - if often have to get up earlier than I like. This excludes getting woken up unexpectedly in the middle of the night, which is known to mess with people's minds.

I tried to use my hourly Anki performance, but it seems very flat, except indeed for a dip a 6 AM, but that could be lack of data (70 samples).

 

Reply
Screwtape's Shortform
Gunnar_Zarncke7d21

Reminds me loosely of The Honest Broker.

Reply1
Gunnar_Zarncke's Shortform
Gunnar_Zarncke7d20

Yes! That's the one. Thank you.

Reply
Gunnar_Zarncke's Shortform
Gunnar_Zarncke7d20

I'm looking for a video of AI gone wrong illustrating AI risk and unusual persuasion. It starts with a hall with blinking computers where an AI voice is manipulating a janitor and it ends with a plane crashing and other emergencies. I think it was made between 2014 and 2018 and linked on LW but I can't google, perplex or o3 it. And ideas?

Reply
On the functional self of LLMs
Gunnar_Zarncke8d40

Are you implying that there is a connection between A Three-Layer Model of LLM Psychology and active inference or do you offer that just as two lenses into LLM identity? If it is the former, can you say more?

Reply
On the functional self of LLMs
Gunnar_Zarncke8d20

There is a looong conversation between Eliezer Yudkowsky and Robin Hanson on X about how LLMs model humans.

Reply1
Lightcone Infrastructure/LessWrong is looking for funding
Gunnar_Zarncke12d20

Thanks, yes that mostly answers it. I got curious when the Buddhist temple thing was mentioned. 11 of 52 weekends not-booked implies a very high utilization, and I'd guess that you had to turn away customers (or at least delay) and it seems you could defer to other locations (though nothing beats Lighthaven of course). 

Reply
Lightcone Infrastructure/LessWrong is looking for funding
Gunnar_Zarncke13d20

the Bay has taken a nosedive in terms of prices since many tech companies have stayed remote post-pandemic

Has this assessment changed since then? I hear many companies are back to on-site.

How well booked is Lighthaven usually? Or rather: Has there been any need for extra capacity?

Reply
Load More
Theory of Mind
10mo
(+250)
Pareto Efficiency
1y
(+52/-52)
Pareto Efficiency
1y
(+52)
Pareto Efficiency
1y
(+392)
Babble and Prune
2y
(+1264)
Has Diagram
2y
(+163)
Simulation
2y
(+9/-10)
Simulation
2y
(+443/-24)
Simulation
2y
(+174/-3)
Simulation
2y
(+646)
Load More
9Hybrid model reveals people act less rationally in complex games, more predictably in simple ones
6d
0
52Project Vend: Can Claude run a small shop?
15d
7
13[Linkpost] The lethal trifecta for AI agents: private data, untrusted content, and external communication
1mo
3
34Unexpected Conscious Entities
2mo
6
13[Linkpost] The value of initiating a pursuit in temporal decision-making
4mo
0
81Mistral Large 2 (123B) seems to exhibit alignment faking
Ω
4mo
Ω
4
156Reducing LLM deception at scale with self-other overlap fine-tuning
Ω
4mo
Ω
42
63RL, but don't do anything I wouldn't do
7mo
5
13[Linkpost] Building Altruistic and Moral AI Agent with Brain-inspired Affective Empathy Mechanisms
8mo
0
7Consciousness As Recursive Reflections
9mo
2
Load More