AI ALIGNMENT FORUMTags
AF

Gears-Level

EditHistorySubscribe
Discussion (0)
Help improve this page
EditHistorySubscribe
Discussion (0)
Help improve this page
Gears-Level
Random Tag
Contributors
2Ruben Bloom
1B Jacobs
1brook

A gears-level model is 'well-constrained' in the sense that there is a strong connection between each of the things you observe-- it would be hard for you to imagine that one of the variables could be different while all of the others remained the same.

Related Tags: Anticipated Experiences, Double-Crux, Empiricism, Falsifiability, Map and Territory


The term gears-level was first described on LW in the post "Gears in Understanding":...

(Read More)

Posts tagged Gears-Level
3
22Toward a New Technical Explanation of Technical Explanation
Abram Demski
6y
2
0
51Attempted Gears Analysis of AGI Intervention Discussion With Eliezer
Zvi
2y
0
1
52Evolution of Modularity
johnswentworth
4y
6
1
34A Case for the Least Forgiving Take On Alignment
Thane Ruthenis
7mo
18
1
15Abstraction, Evolution and Gears
johnswentworth
3y
4
0
67interpreting GPT: the logit lens
nostalgebraist
3y
14
1
38Current themes in mechanistic interpretability research
Lee Sharkey, Sid Black, Beren Millidge
1y
2
1
34Decision Transformer Interpretability
Joseph Isaac Bloom, Paul Colognese
10mo
6
0
11Beware of black boxes in AI alignment research
Vladimir Slepnev
6y
0
1
17Value Formation: An Overarching Model
Thane Ruthenis
1y
10
Add Posts