Amanda Askell
Philosopher and AI alignment researcher at Anthropic; leads work on Claude's character and constitutional AI design.
- Role
- researcher
- Currently at
- Anthropic
- twitter.com/AmandaAskell
- Scholar
- scholar.google.com/citations
- Papers
- 8
Cite
Notes
Only stored in your browser.
Authored papers
8Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
arXiv 2024
Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
preprint
Training language models to follow instructions with human feedback
NeurIPS
Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
TMLR
Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned
arXiv 2022
Discovering Language Model Behaviors with Model-Written Evaluations
arXiv 2022
Constitutional AI: Harmlessness from AI Feedback
arXiv 2022
Learning Transferable Visual Models From Natural Language Supervision
arXiv 2021
Affiliations
Previously
Frequent co-authors
10from 8 papers
Deep Ganguli
researcher
Jared Kaplan
co-founder / Chief Science Officer
Yuntao Bai
researcher
Danny Hernandez
researcher
Jack Clark
founder
Jackson Kernion
researcher
Kamal Ndousse
researcher
Nova DasSarma
researcher
Shauna Kravec
researcher
Andy Jones
researcher