AI ALIGNMENT FORUM
AF

Anthropic (org)

Edited by Dakara, Multicore, Ruby, ryan_greenblatt last updated 31st Dec 2024

Anthropic is an AI company based in San Francisco. The company is known for developing the Claude AI family and publishing research on AI safety.

Not to be confused with anthropics.

1

1

Posts tagged Anthropic (org)

63Anthropic's Core Views on AI Safety

Zac Hatfield-Dodds

2y

13

49Why I'm joining Anthropic

3y

2

33Toy Models of Superposition

3y

2

29Concrete Reasons for Hope about AI

Zac Hatfield-Dodds

3y

0

67Transformer Circuits

4y

3

110Towards Monosemanticity: Decomposing Language Models With Dictionary Learning

Zac Hatfield-Dodds

2y

12

95Introducing Alignment Stress-Testing at Anthropic

2y

19

69EIS XIII: Reflections on Anthropic’s SAE Research Circa May 2024

1y

7

42Anthropic: Three Sketches of ASL-4 Safety Case Components

Zach Stein-Perlman

10mo

18

45Anthropic's Responsible Scaling Policy & Long-Term Benefit Trust

Zac Hatfield-Dodds

2y

6

45Dario Amodei’s prepared remarks from the UK AI Safety Summit, on Anthropic’s Responsible Scaling Policy

Zac Hatfield-Dodds

2y

0

30Paper: The Capacity for Moral Self-Correction in Large Language Models (Anthropic)

3y

0

30Putting up Bumpers

4mo

7

18How do new models from OpenAI, DeepMind and Anthropic perform on TruthfulQA?

4y

1

19Anthropic's updated Responsible Scaling Policy

Zac Hatfield-Dodds

11mo

0

Load More (15/21)

Add Posts