This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
Tags
LW
Login
Transformers
Edit
History
Subscribe
Discussion
(0)
Help improve this page
Edit
History
Subscribe
Discussion
(0)
Help improve this page
Transformers
Random Tag
Contributors
Posts tagged
Transformers
Most Relevant
5
35
Striking Implications for Learning Theory, Interpretability — and Safety?
RogerDearnaley
4mo
4
3
125
How LLMs are and are not myopic
Ω
janus
9mo
Ω
14
2
197
Modern Transformers are AGI, and Human-Level
Ω
abramdemski
1mo
Ω
89
2
86
Google's PaLM-E: An Embodied Multimodal Language Model
SandXbox
1y
7
2
72
Residual stream norms grow exponentially over the forward pass
Ω
StefanHex
,
TurnTrout
1y
Ω
24
2
62
Tracr: Compiled Transformers as a Laboratory for Interpretability | DeepMind
Ω
DragonGod
1y
Ω
12
2
54
Concrete Steps to Get Started in Transformer Mechanistic Interpretability
Ω
Neel Nanda
1y
Ω
7
2
53
How fast can we perform a forward pass?
jsteinhardt
2y
9
2
33
AGI will be made of heterogeneous components, Transformer and Selective SSM blocks will be among them
Ω
Roman Leventov
4mo
Ω
9
2
27
How Do Induction Heads Actually Work in Transformers With Finite Capacity?
Fabien Roger
1y
0
1
334
Transformers Represent Belief State Geometry in their Residual Stream
Ω
Adam Shai
16h
Ω
67
1
81
An Analogy for Understanding Transformers
CallumMcDougall
1y
5
1
76
Attention SAEs Scale to GPT-2 Small
Ω
Connor Kissane
,
robertzk
,
Arthur Conmy
,
Neel Nanda
3mo
Ω
4
1
53
Skepticism About DeepMind's "Grandmaster-Level" Chess Without Search
Arjun Panickssery
3mo
13
1
47
Brief Notes on Transformers
Ω
Adam Jermyn
2y
Ω
3