Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
95 posts
Interpretability (ML & AI)
AI Success Models
Lottery Ticket Hypothesis
Conservatism (AI)
Market making (AI safety technique)
Verification
28 posts
Myopia
Self Fulfilling/Refuting Prophecies
Luck
16
An Open Agency Architecture for Safe Transformative AI
davidad
11h
11
140
How "Discovering Latent Knowledge in Language Models Without Supervision" Fits Into a Broader Alignment Scheme
Collin
5d
18
21
Paper: Transformers learn in-context by gradient descent
LawrenceC
4d
11
250
The Plan - 2022 Update
johnswentworth
19d
33
26
The limited upside of interpretability
Peter S. Park
1mo
11
235
The Singular Value Decompositions of Transformer Weight Matrices are Highly Interpretable
beren
22d
27
56
Multi-Component Learning and S-Curves
Adam Jermyn
20d
24
78
By Default, GPTs Think In Plain Sight
Fabien Roger
1mo
16
45
"Cars and Elephants": a handwavy argument/analogy against mechanistic interpretability
David Scott Krueger (formerly: capybaralet)
1mo
25
151
Re-Examining LayerNorm
Eric Winsor
19d
8
35
Extracting and Evaluating Causal Direction in LLMs' Activations
Fabien Roger
6d
2
31
A Walkthrough of Interpretability in the Wild (w/ authors Kevin Wang, Arthur Conmy & Alexandre Variengien)
Neel Nanda
1mo
15
85
Real-Time Research Recording: Can a Transformer Re-Derive Positional Info?
Neel Nanda
1mo
14
56
A Mystery About High Dimensional Concept Encoding
Fabien Roger
1mo
13
34
Steering Behaviour: Testing for (Non-)Myopia in Language Models
Evan R. Murphy
15d
16
15
Generative, Episodic Objectives for Safe AI
Michael Glass
2mo
3
53
LCDT, A Myopic Decision Theory
adamShimi
1y
51
32
Random Thoughts on Predict-O-Matic
abramdemski
3y
3
21
An example of self-fulfilling spurious proofs in UDT
cousin_it
10y
43
6
Luck II: Expecting White Swans
fowlertm
9y
87
56
Partial Agency
abramdemski
3y
18
79
The Credit Assignment Problem
abramdemski
3y
40
47
Why GPT wants to mesa-optimize & how we might change this
John_Maxwell
2y
32
46
Towards a mechanistic understanding of corrigibility
evhub
3y
26
23
Self-Supervised Learning and AGI Safety
Steven Byrnes
3y
9
48
Acceptability Verification: A Research Agenda
David Udell
5mo
0
47
Arguments against myopic training
Richard_Ngo
2y
39
13
Encouragement to Instill Confidence?
Zvi
10mo
5