Publications: Amanda Askell

← Back to Author Profile

  • Affiliation: Anthropic
  • Author ID: NYOJzM4AAAAJ
  • Cited publications: 67

Download CSV for Amanda Askell

Title Year Citations Score
Language models are few-shot learners
Advances in neural information processing systems 33, 1877-1901, 2020
View on Google Scholar ↗
2020 98447 100.0%
Language Models are Few-Shot Learners (GPT-3)
2026
View on Google Scholar ↗
2026 1890 100.0%
Learning transferable visual models from natural language supervision
International conference on machine learning, 8748-8763, 2021
View on Google Scholar ↗
2021 78192 100.0%
Training language models to follow instructions with human feedback
Advances in neural information processing systems 35, 27730-27744, 2022
View on Google Scholar ↗
2022 34129 100.0%
Constitutional ai: Harmlessness from ai feedback
arXiv preprint arXiv:2212.08073, 2022
View on Google Scholar ↗
2022 5751 100.0%
Training a helpful and harmless assistant with reinforcement learning from human feedback
arXiv preprint arXiv:2204.05862, 2022
View on Google Scholar ↗
2022 4998 100.0%
Beyond the imitation game: Quantifying and extrapolating the capabilities of language models
arXiv preprint arXiv:2206.04615, 2022
View on Google Scholar ↗
2022 3201 99.9%
Towards understanding sycophancy in language models
International Conference on Learning Representations 2024, 110-144, 2024
View on Google Scholar ↗
2024 2799 99.9%
Language models (mostly) know what they know
arXiv preprint arXiv:2207.05221, 2022
View on Google Scholar ↗
2022 2377 99.9%
Towards monosemanticity: Decomposing language models with dictionary learning
Transformer Circuits Thread 2 (5), 6, 2023
View on Google Scholar ↗
2023 1939 99.9%
In-context learning and induction heads
arXiv preprint arXiv:2209.11895, 2022
View on Google Scholar ↗
2022 1581 99.8%
Discovering language model behaviors with model-written evaluations
Findings of the association for computational linguistics: ACL 2023, 13387-13434, 2023
View on Google Scholar ↗
2023 1502 99.8%
Red teaming language models to reduce harms: Methods, scaling behaviors, and lessons learned
arXiv preprint arXiv:2209.07858, 2022
View on Google Scholar ↗
2022 1447 99.8%
A mathematical framework for transformer circuits
Transformer Circuits Thread 1 (1), 12, 2021
View on Google Scholar ↗
2021 1681 99.8%
Sleeper agents: Training deceptive llms that persist through safety training
arXiv preprint arXiv:2401.05566, 2024
View on Google Scholar ↗
2024 819 99.7%
A general language assistant as a laboratory for alignment
arXiv preprint arXiv:2112.00861, 2021
View on Google Scholar ↗
2021 1075 99.6%
Towards measuring the representation of subjective global opinions in language models
arXiv preprint arXiv:2306.16388, 2023
View on Google Scholar ↗
2023 608 99.6%
Dawn Drain
Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Scott Johnston, Andy …, 2022
View on Google Scholar ↗
2022 634 99.5%
Predictability and surprise in large generative models
Proceedings of the 2022 ACM conference on fairness, accountability, and …, 2022
View on Google Scholar ↗
2022 592 99.4%
Dawn Drain
Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Andy Jones, Jackson …, 2021
View on Google Scholar ↗
2021 758 99.4%
Release strategies and the social impacts of language models
arXiv preprint arXiv:1908.09203, 2019
View on Google Scholar ↗
2019 1176 99.3%
Toward trustworthy AI development: mechanisms for supporting verifiable claims
arXiv preprint arXiv:2004.07213, 2020
View on Google Scholar ↗
2020 1039 99.3%
Constitutional classifiers: Defending against universal jailbreaks across thousands of hours of red teaming
arXiv preprint arXiv:2501.18837, 2025
View on Google Scholar ↗
2025 241 99.2%
The capacity for moral self-correction in large language models
arXiv preprint arXiv:2302.07459, 2023
View on Google Scholar ↗
2023 306 99.1%
Measuring progress on scalable oversight for large language models
arXiv preprint arXiv:2211.03540, 2022
View on Google Scholar ↗
2022 352 99.0%
Towards understanding sycophancy in language models, 2023
URL https://arxiv. org/abs/2310.13548 2 (5), 32, 2023
View on Google Scholar ↗
2023 256 98.9%
Evaluating and mitigating discrimination in language model decisions
arXiv preprint arXiv:2312.03689, 2023
View on Google Scholar ↗
2023 221 98.7%
International conference on machine learning
International Conference on Machine Learning, 8748-8763, 2021
View on Google Scholar ↗
2021 430 98.6%
Training a helpful and harmless assistant with reinforcement learning from human feedback, 2022
URL https://arxiv. org/abs/2204.05862, 2025
View on Google Scholar ↗
2025 147 98.6%
Dawn Drain
Ethan Perez, Nicholas Schiefer, Zac Hatfield-Dodds, Nova DasSarma, Eli Tran …, 2022
View on Google Scholar ↗
2022 235 98.3%
Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv
arXiv preprint arXiv:2204.05862, 2022
View on Google Scholar ↗
2022 215 98.1%
Dawn Drain
Stanislav Fort, Deep Ganguli, Tom Henighan, Nicholas Joseph, Saurav Kadavath …, 2022
View on Google Scholar ↗
2022 211 98.0%
Dawn Drain, Deep Ganguli, Tom Henighan, Andy Jones, Nicholas Joseph, Ben Mann, Nova DasSarma, et al. 2021. A general language assistant as a laboratory for alignment
arXiv preprint arXiv:2112.00861, 2021
View on Google Scholar ↗
2021 249 97.3%
Dawn Drain, Ethan Perez, Nicholas Schiefer, Zac Hatfield-Dodds, Nova DasSarma, Eli Tran-Johnson, and 1 others. 2022. Language models (mostly) know what they know
arXiv preprint arXiv:2207.05221, 2022
View on Google Scholar ↗
2022 159 97.2%
Constitutional ai: Harmlessness from ai feedback. arXiv 2022
arXiv preprint arXiv:2212.08073, 2022
View on Google Scholar ↗
2022 116 95.9%
Towards monosemanticity: Decomposing language models with dictionary learning, 2023
URL https://transformer-circuits. pub/2023/monosemantic-features/index. html, 9, 2025
View on Google Scholar ↗
2025 62 95.9%
Language models (mostly) know what they know (2022)
URL https://arxiv. org/abs/2207.05221 5, 2023
View on Google Scholar ↗
2023 79 95.8%
Specific versus general principles for constitutional ai
arXiv preprint arXiv:2310.13798, 2023
View on Google Scholar ↗
2023 73 95.4%
January 18-24). Learning transferable visual models from natural language supervision. Proceedings of the International Conference on Machine Learning, Virtual
2021
View on Google Scholar ↗
2021 137 94.3%
Dawn Drain, Deep Ganguli, Tom Henighan, Andy Jones, Nicholas Joseph
arXiv preprint arXiv:2112.00861, 2021
View on Google Scholar ↗
2021 133 94.1%
Dawn Drain
Deep Ganguli, Tom Henighan, Andy Jones, Nicholas Joseph, Ben Mann, Nova …, 2021
View on Google Scholar ↗
2021 123 93.6%
Training language models to follow instructions with human feedback, 2022. doi: 10.48550
arXiv preprint ARXIV.2203.02155, 2022
View on Google Scholar ↗
2022 76 93.2%
Learning transferable visual models from natural language supervision
Computer Science, 2021
View on Google Scholar ↗
2021 111 92.8%
Larochelle H., Ranzato M., Hadsell R., Balcan MF, Lin H
Language models are few-shot learners. Advances in neural information …, 2020
View on Google Scholar ↗
2020 146 92.2%
Beyond the imitation game: Quantifying and extrapolating the capabilities of language models, 2023
URL https://arxiv. org/abs/2206.04615, 2022
View on Google Scholar ↗
2022 67 92.1%
Training language models to follow instructions with human feedback. arXiv. 2022. DOI: 10.48550
arXiv preprint arXiv.2203.02155 2, 2022
View on Google Scholar ↗
2022 67 92.1%
Softmax linear units
Transformer Circuits Thread, 95, 2022
View on Google Scholar ↗
2022 54 89.9%
Red teaming language models to reduce harms: Methods, scaling behaviors, and lessons learned. arXiv
arXiv preprint arXiv:2209.07858, 2022
View on Google Scholar ↗
2022 52 89.5%
A general language assistant as a laboratory for alignment. arXiv
arXiv preprint arXiv:2112.00861, 2021
View on Google Scholar ↗
2021 77 89.0%
The role of cooperation in responsible AI development
arXiv preprint arXiv:1907.04534, 2019
View on Google Scholar ↗
2019 123 88.7%
Learning transferable visual models from natural language supervision, 139
2021
View on Google Scholar ↗
2021 71 87.9%
Nova Das-Sarma
Dawn Drain, Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Andy Jones …, 2021
View on Google Scholar ↗
2021 67 87.1%
AI safety needs social scientists
Distill 4 (2), e14, 2019
View on Google Scholar ↗
2019 108 87.0%
& Lowe, R.(2022)
Training language models to follow instructions with human feedback 5, 88, 2022
View on Google Scholar ↗
2022 41 86.5%
Language models are few-shot learners. CoRR abs/2005.14165 (2020)
2005
View on Google Scholar ↗
2005 195 82.7%
& Amodei, D.(2020)
Language models are few-shot learners 15, 1877-1901, 2005
View on Google Scholar ↗
2005 191 82.4%
teusz Litwin
M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A …, 2020
View on Google Scholar ↗
2020 64 81.3%
Pareto principles in infinite ethics
New York University, 2018
View on Google Scholar ↗
2018 56 71.3%
The moral inefficacy of carbon offsetting
Australasian Journal of Philosophy 102 (4), 795-813, 2024
View on Google Scholar ↗
2024 7 66.6%
Longtermist Myopia
Essays on Longtermism: Present Action for the Distant Future, 17-49, 2025
View on Google Scholar ↗
2025 6 65.5%
Language models are few-shot learners. arXiv. org
2005
View on Google Scholar ↗
2005 66 61.6%
Rewon Child, et al. 2020. Language models are few-shot learners
Advances in neural information processing systems. https://doi. org/10.48550 …, 2005
View on Google Scholar ↗
2005 55 57.7%
Evidence Neutrality and the Moral Value of Information
Effective Altruism: Philosophical Issues, 37, 2019
View on Google Scholar ↗
2019 13 38.4%
AI safety needs social scientists. Distill, 4 (2), e14
2019
View on Google Scholar ↗
2019 7 26.6%
Rewon Child, Aditya Ramesh, Daniel Ziegler, Je rey Wu, Clemens Winter, Chris Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Sco Gray
Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford …, 2020
View on Google Scholar ↗
2020 3 16.2%
Prudential objections to atheism
A companion to atheism and philosophy, 506-520, 2019
View on Google Scholar ↗
2019 3 15.6%
Epistemic Consequentialism and Epistemic Enkrasia
Epistemic Consequentialism, 290-309, 2018
View on Google Scholar ↗
2018 2 11.3%