Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
39 posts
Language Models
Scaling Laws
Definitions
PaLM
Prompt Engineering
Robotics
Exploratory Engineering
Simulacrum Levels
16 posts
Agency
Deconfusion
Tool AI
Simulation Hypothesis
Philosophy of Language
Astronomical Waste
Carving / Clustering Reality
494
chinchilla's wild implications
nostalgebraist
4mo
114
191
Announcing the Inverse Scaling Prize ($250k Prize Pool)
Ethan Perez
5mo
14
165
Language models seem to be much better than humans at next-token prediction
Buck
4mo
56
140
Who models the models that model models? An exploration of GPT-3's in-context model fitting ability
Lovre
6mo
14
127
Transformer Circuits
evhub
12mo
4
123
The case for becoming a black-box investigator of language models
Buck
7mo
19
116
RL with KL penalties is better seen as Bayesian inference
Tomek Korbak
6mo
15
101
Causal confusion as an argument against the scaling hypothesis
RobertKirk
6mo
30
101
Inverse Scaling Prize: Round 1 Winners
Ethan Perez
2mo
16
100
Testing PaLM prompts on GPT3
Yitz
8mo
15
89
Help ARC evaluate capabilities of current language models (still need people)
Beth Barnes
5mo
6
80
Thoughts on the Alignment Implications of Scaling Language Models
leogao
1y
11
75
Language Model Alignment Research Internships
Ethan Perez
1y
1
70
Gears-Level Mental Models of Transformer Interpretability
KevinRoWang
8mo
4
759
Simulators
janus
3mo
103
130
Beyond Astronomical Waste
Wei_Dai
4y
41
79
Discovering Agents
zac_kenton
4mo
8
66
Looking Deeper at Deconfusion
adamShimi
1y
13
62
LOVE in a simbox is all you need
jacob_cannell
2mo
69
51
Vingean Agency
abramdemski
3mo
13
37
Why agents are powerful
Daniel Kokotajlo
6mo
7
26
Agency: What it is and why it matters
Daniel Kokotajlo
1y
2
23
[Intro to brain-like-AGI safety] 11. Safety ≠ alignment (but they’re close!)
Steven Byrnes
8mo
1
22
Some reasons why a predictor wants to be a consequentialist
Lauro Langosco
8mo
16
19
Why not tool AI?
smithee
3y
10
18
Beware over-use of the agent model
Alex Flint
1y
10
14
Pitfalls of the agent model
Alex Flint
1y
4
9
An Agent is a Worldline in Tegmark V
komponisto
4y
12