This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
AI ALIGNMENT FORUM
AF
Login
Home
Library
Questions
All Posts
About
Top Questions
14
What specific dangers arise when asking GPT-N to write an Alignment Forum post?
Q
Matthew Barnett
,
orthonormal
15d
Q
10
21
To what extent is GPT-3 capable of reasoning?
Q
Alex Turner
,
Daniel Kokotajlo
22d
Q
16
21
How should AI debate be judged?
Q
Abram Demski
,
Paul Christiano
1mo
Q
27
18
Why is pseudo-alignment "worse" than other ways ML can fail to generalize?
Q
nostalgebraist
,
Evan Hubinger
24d
Q
6
26
How "honest" is GPT-3?
Q
Abram Demski
,
gwern
1mo
Q
4
Recent Activity
6
What are the most important papers/post/resources to read to understand more of GPT-3?
Q
Adam Shimi
9d
Q
4
14
What specific dangers arise when asking GPT-N to write an Alignment Forum post?
Q
Matthew Barnett
,
orthonormal
15d
Q
10
11
What if memes are common in highly capable minds?
Q
Daniel Kokotajlo
12d
Q
2
21
To what extent is GPT-3 capable of reasoning?
Q
Alex Turner
,
Daniel Kokotajlo
22d
Q
16
10
Probability that other architectures will scale as well as Transformers?
Q
Daniel Kokotajlo
,
Rohin Shah
14d
Q
3
11
To what extent are the scaling properties of Transformer networks exceptional?
Q
Abram Demski
14d
Q
1
12
What happens to variance as neural network training is scaled? What does it imply about "lottery tickets"?
Q
Abram Demski
14d
Q
1
6
Does the lottery ticket hypothesis suggest the scaling hypothesis?
Q
Daniel Kokotajlo
14d
Q
2
21
How should AI debate be judged?
Q
Abram Demski
,
Paul Christiano
1mo
Q
27
5
What is the alternative to intent alignment called?
Q
Richard Ngo
3mo
Q
4
18
Why is pseudo-alignment "worse" than other ways ML can fail to generalize?
Q
nostalgebraist
,
Evan Hubinger
24d
Q
6
26
How "honest" is GPT-3?
Q
Abram Demski
,
gwern
1mo
Q
4
