Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
153 posts
Interpretability (ML & AI)
GPT
Machine Learning (ML)
DeepMind
OpenAI
Truth, Semantics, & Meaning
AI Success Models
Bounties & Prizes (active)
AI-assisted Alignment
Lottery Ticket Hypothesis
Computer Science
Honesty
168 posts
Conjecture (org)
Oracle AI
Myopia
Language Models
Refine
Deconfusion
Agency
AI Boxing (Containment)
Deceptive Alignment
Scaling Laws
Deception
Acausal Trade
13
An Open Agency Architecture for Safe Transformative AI
davidad
11h
11
123
How "Discovering Latent Knowledge in Language Models Without Supervision" Fits Into a Broader Alignment Scheme
Collin
5d
18
265
A challenge for AGI organizations, and a challenge for readers
Rob Bensinger
19d
30
211
The Plan - 2022 Update
johnswentworth
19d
33
59
Predicting GPU performance
Marius Hobbhahn
6d
24
26
Paper: Transformers learn in-context by gradient descent
LawrenceC
4d
11
93
[Link] Why I’m optimistic about OpenAI’s alignment approach
janleike
15d
13
47
Reframing inner alignment
davidad
9d
13
99
Re-Examining LayerNorm
Eric Winsor
19d
8
22
Extracting and Evaluating Causal Direction in LLMs' Activations
Fabien Roger
6d
2
26
An exploration of GPT-2's embedding weights
Adam Scherlis
7d
2
31
[ASoT] Natural abstractions and AlphaZero
Ulisse Mini
10d
1
57
Multi-Component Learning and S-Curves
Adam Jermyn
20d
24
364
DeepMind alignment team opinions on AGI ruin arguments
Vika
4mo
34
27
Discovering Language Model Behaviors with Model-Written Evaluations
evhub
4h
3
55
Proper scoring rules don’t guarantee predicting fixed points
Johannes_Treutlein
4d
2
29
Take 11: "Aligning language models" should be weirder.
Charlie Steiner
2d
0
80
[Interim research report] Taking features out of superposition with sparse autoencoders
Lee Sharkey
7d
10
159
The Singular Value Decompositions of Transformer Weight Matrices are Highly Interpretable
beren
22d
27
183
Conjecture: a retrospective after 8 months of work
Connor Leahy
27d
9
35
Side-channels: input versus output
davidad
8d
9
213
Mysteries of mode collapse
janus
1mo
35
103
What I Learned Running Refine
adamShimi
26d
5
472
Simulators
janus
3mo
103
85
Conjecture Second Hiring Round
Connor Leahy
27d
0
64
Searching for Search
NicholasKees
22d
6
37
Steering Behaviour: Testing for (Non-)Myopia in Language Models
Evan R. Murphy
15d
16
82
Current themes in mechanistic interpretability research
Lee Sharkey
1mo
3