Publications: Dario Amodei
| 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 | 86239 | 100.0% |
|
Language models are unsupervised multitask learners
OpenAI blog 1 (8), 9, 2019 View on Google Scholar ↗ |
2019 | 25493 | 100.0% |
|
Scaling laws for neural language models
arXiv preprint arXiv:2001.08361, 2020 View on Google Scholar ↗ |
2020 | 9630 | 100.0% |
|
Constitutional ai: Harmlessness from ai feedback
arXiv preprint arXiv:2212.08073, 2022 View on Google Scholar ↗ |
2022 | 5746 | 100.0% |
|
Evaluating large language models trained on code
arXiv preprint arXiv:2107.03374, 2021 View on Google Scholar ↗ |
2021 | 12636 | 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 | 4978 | 100.0% |
|
Language Models are Few-Shot Learners (GPT-3)
2026 View on Google Scholar ↗ |
2026 | 1883 | 99.9% |
|
Deep reinforcement learning from human preferences
Advances in neural information processing systems 30, 2017 View on Google Scholar ↗ |
2017 | 9065 | 99.9% |
|
Language models (mostly) know what they know
arXiv preprint arXiv:2207.05221, 2022 View on Google Scholar ↗ |
2022 | 2335 | 99.9% |
|
Learning to summarize with human feedback
Advances in neural information processing systems 33, 3008-3021, 2020 View on Google Scholar ↗ |
2020 | 4377 | 99.9% |
|
Fine-tuning language models from human preferences
arXiv preprint arXiv:1909.08593, 2019 View on Google Scholar ↗ |
2019 | 3476 | 99.9% |
|
In-context learning and induction heads
arXiv preprint arXiv:2209.11895, 2022 View on Google Scholar ↗ |
2022 | 1565 | 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 | 1475 | 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 | 1438 | 99.8% |
|
Concrete problems in AI safety
arXiv preprint arXiv:1606.06565, 2016 View on Google Scholar ↗ |
2016 | 6625 | 99.8% |
|
A mathematical framework for transformer circuits
Transformer Circuits Thread 1 (1), 12, 2021 View on Google Scholar ↗ |
2021 | 1653 | 99.8% |
|
Toy models of superposition
arXiv preprint arXiv:2209.10652, 2022 View on Google Scholar ↗ |
2022 | 1108 | 99.7% |
|
Deep speech 2: End-to-end speech recognition in english and mandarin
International conference on machine learning, 173-182, 2016 View on Google Scholar ↗ |
2016 | 4543 | 99.7% |
|
A cross-platform toolkit for mass spectrometry and proteomics
Nature biotechnology 30 (10), 918-920, 2012 View on Google Scholar ↗ |
2012 | 4993 | 99.7% |
|
Language models are unsupervised multitask learners (2019)
2025 View on Google Scholar ↗ |
2025 | 572 | 99.6% |
|
A general language assistant as a laboratory for alignment
arXiv preprint arXiv:2112.00861, 2021 View on Google Scholar ↗ |
2021 | 1072 | 99.6% |
|
The malicious use of artificial intelligence: Forecasting, prevention, and mitigation
arXiv preprint arXiv:1802.07228, 2018 View on Google Scholar ↗ |
2018 | 2335 | 99.6% |
|
Which economic tasks are performed with ai? evidence from millions of claude conversations
arXiv preprint arXiv:2503.04761, 2025 View on Google Scholar ↗ |
2025 | 333 | 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 | 589 | 99.4% |
|
The capacity for moral self-correction in large language models
arXiv preprint arXiv:2302.07459, 2023 View on Google Scholar ↗ |
2023 | 304 | 99.1% |
|
Benchmarking safe exploration in deep reinforcement learning
arXiv preprint arXiv:1910.01708 7 (1), 2, 2019 View on Google Scholar ↗ |
2019 | 923 | 99.0% |
|
Scaling laws for autoregressive generative modeling
arXiv preprint arXiv:2010.14701, 2020 View on Google Scholar ↗ |
2020 | 780 | 99.0% |
|
Measuring progress on scalable oversight for large language models
arXiv preprint arXiv:2211.03540, 2022 View on Google Scholar ↗ |
2022 | 349 | 98.9% |
|
Language models are unsupervised multitask learners. OpenAI Blog (2019)
URL: https://d4mucfpksywv. cloudfront. net/better-language-models/language …, 2024 View on Google Scholar ↗ |
2024 | 178 | 98.7% |
|
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 | 144 | 98.5% |
|
Evaluating large language models trained on code. arXiv 2021
arXiv preprint arXiv:2107.03374 10, 2021 View on Google Scholar ↗ |
2021 | 379 | 98.4% |
|
AI safety via debate
arXiv preprint arXiv:1805.00899, 2018 View on Google Scholar ↗ |
2018 | 737 | 98.2% |
|
Reward learning from human preferences and demonstrations in atari
Advances in neural information processing systems 31, 2018 View on Google Scholar ↗ |
2018 | 716 | 98.1% |
|
Machines of loving grace: How AI could transform the world for the better
Dario Amodei 11, 2024 View on Google Scholar ↗ |
2024 | 127 | 98.0% |
|
AI and Compute
2018 View on Google Scholar ↗ |
2018 | 673 | 97.9% |
|
Machines of loving grace
Essay on Anthropic CEO’s website, 2024 View on Google Scholar ↗ |
2024 | 119 | 97.8% |
|
The adolescence of technology
DarioAmodei. com 15, 2026 View on Google Scholar ↗ |
2026 | 59 | 97.3% |
|
Better language models and their implications
OpenAI blog 1 (2), 2019 View on Google Scholar ↗ |
2019 | 392 | 97.1% |
|
An empirical model of large-batch training
arXiv preprint arXiv:1812.06162, 2018 View on Google Scholar ↗ |
2018 | 496 | 97.1% |
|
Characterizing deformability and surface friction of cancer cells
Proceedings of the National Academy of Sciences 110 (19), 7580-7585, 2013 View on Google Scholar ↗ |
2013 | 491 | 95.8% |
|
Language models (mostly) know what they know (2022)
URL https://arxiv. org/abs/2207.05221 5, 2023 View on Google Scholar ↗ |
2023 | 78 | 95.8% |
|
Constitutional ai: Harmlessness from ai feedback. arXiv 2022
arXiv preprint arXiv:2212.08073, 2022 View on Google Scholar ↗ |
2022 | 112 | 95.7% |
|
Building high-quality assay libraries for targeted analysis of SWATH MS data
Nature protocols 10 (3), 426-441, 2015 View on Google Scholar ↗ |
2015 | 399 | 95.3% |
|
Scaling laws and interpretability of learning from repeated data
arXiv preprint arXiv:2205.10487, 2022 View on Google Scholar ↗ |
2022 | 102 | 95.2% |
|
The urgency of interpretability
Dario Amodei 1, 2025 View on Google Scholar ↗ |
2025 | 48 | 94.6% |
|
Language models are few-shot learners
arXiv preprint arXiv:2005.14165, 2020 View on Google Scholar ↗ |
2020 | 193 | 94.4% |
|
Variational option discovery algorithms
arXiv preprint arXiv:1807.10299, 2018 View on Google Scholar ↗ |
2018 | 277 | 94.2% |
|
Physical principles for scalable neural recording
Frontiers in computational neuroscience 7, 137, 2013 View on Google Scholar ↗ |
2013 | 371 | 94.1% |
|
Searching for collective behavior in a large network of sensory neurons
PLoS computational biology 10 (1), e1003408, 2014 View on Google Scholar ↗ |
2014 | 343 | 94.1% |
|
Thermodynamics and signatures of criticality in a network of neurons
Proceedings of the National Academy of Sciences 112 (37), 11508-11513, 2015 View on Google Scholar ↗ |
2015 | 316 | 93.7% |
|
Improving precursor selectivity in data-independent acquisition using overlapping windows
Journal of the American Society for Mass Spectrometry 30 (4), 669-684, 2019 View on Google Scholar ↗ |
2019 | 195 | 93.3% |
|
Supervising strong learners by amplifying weak experts
arXiv preprint arXiv:1810.08575, 2018 View on Google Scholar ↗ |
2018 | 242 | 93.2% |
|
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 | 145 | 92.1% |
|
Deployed end-to-end speech recognition
US Patent 10,319,374, 2019 View on Google Scholar ↗ |
2019 | 157 | 91.4% |
|
End-to-end speech recognition
US Patent 10,332,509, 2019 View on Google Scholar ↗ |
2019 | 150 | 90.9% |
|
On DeepSeek and export controls
SuperIntelligence-Robotics-Safety & Alignment 2 (1), 2025 View on Google Scholar ↗ |
2025 | 29 | 90.8% |
|
Faulty reward functions in the wild
Internet: https://blog. openai. com/faulty-reward-functions 40, 53, 2016 View on Google Scholar ↗ |
2016 | 209 | 90.8% |
|
A mathematical framework for transformer circuits, 2021
URL https://transformer-circuits. pub/2021/framework/index. html, 2023 View on Google Scholar ↗ |
2023 | 35 | 90.1% |
|
Softmax linear units
Transformer Circuits Thread, 95, 2022 View on Google Scholar ↗ |
2022 | 54 | 89.9% |
|
Scaling laws for neural language models. arXiv 2020
arXiv preprint arXiv:2001.08361 10, 2001 View on Google Scholar ↗ |
2001 | 359 | 89.6% |
|
A general language assistant as a laboratory for alignment. arXiv
arXiv preprint arXiv:2112.00861, 2021 View on Google Scholar ↗ |
2021 | 76 | 88.8% |
|
Fine-tuning language models from human preferences. arXiv
arXiv preprint arXiv:1909.08593 10, 2019 View on Google Scholar ↗ |
2019 | 123 | 88.7% |
|
Evaluating large language models trained on code. arXiv preprint arXiv: 210703374
2021 View on Google Scholar ↗ |
2021 | 73 | 88.3% |
|
Mapping a complete neural population in the retina
The Journal of Neuroscience 32 (43), 14859-14873, 2012 View on Google Scholar ↗ |
2012 | 211 | 87.8% |
|
Evaluating large language models trained on code. CoRR abs/2107.03374 (2021)
arXiv preprint arXiv:2107.03374, 2021 View on Google Scholar ↗ |
2021 | 70 | 87.8% |
|
Red teaming language models to reduce harms: Methods, scaling behaviors, and lessons learned, 2022
URL https://arxiv. org/abs/2209.07858, 2022 View on Google Scholar ↗ |
2022 | 44 | 87.5% |
|
Learning a natural language interface with neural programmer
arXiv preprint arXiv:1611.08945, 2016 View on Google Scholar ↗ |
2016 | 151 | 86.8% |
|
The simplest maximum entropy model for collective behavior in a neural network
Journal of Statistical Mechanics: Theory and Experiment 2013 (03), P03011, 2013 View on Google Scholar ↗ |
2013 | 185 | 86.7% |
|
Statement on AI risk
Center for AI safety 31 (5), 2023, 2023 View on Google Scholar ↗ |
2023 | 24 | 85.6% |
|
Concrete problems in AI safety. arXiv 2016
arXiv preprint arXiv:1606.06565, 2016 View on Google Scholar ↗ |
2016 | 124 | 83.7% |
|
Language models are few-shot learners. CoRR abs/2005.14165 (2020)
2005 View on Google Scholar ↗ |
2005 | 195 | 82.7% |
|
Red Teaming Language Models to Reduce Harms: Methods
Scaling Behaviors, and Lessons Learned 2, 2022 View on Google Scholar ↗ |
2022 | 31 | 82.0% |
|
AI and compute, 2018
URL https://openai. com/blog/ai-and-compute 4, 2018 View on Google Scholar ↗ |
2018 | 95 | 82.0% |
|
Better language models and their implications. OpenAI Blog (2019)
2024 View on Google Scholar ↗ |
2024 | 15 | 81.6% |
|
Better language models and their implications. 2019
OpenAI Blog https://openai com/blog/better-language-models, 2025 View on Google Scholar ↗ |
2025 | 14 | 81.5% |
|
Discovering language model behaviors with model-written evaluations. arXiv
arXiv preprint arXiv:2212.09251, 2022 View on Google Scholar ↗ |
2022 | 30 | 81.5% |
|
Identification of a Set of Conserved Eukaryotic Internal Retention Time Standards for Data-independent Acquisition Mass Spectrometry*[S]
Molecular & Cellular Proteomics 14 (10), 2800-2813, 2015 View on Google Scholar ↗ |
2015 | 115 | 80.9% |
|
Learning to summarize from human feedback, 2022
URL https://arxiv. org/abs, 2009 View on Google Scholar ↗ |
2009 | 108 | 73.7% |
|
Faulty reward functions in the wild
URL: https://blog. openai. com/faulty-reward-functions, 2016 View on Google Scholar ↗ |
2016 | 63 | 69.6% |
|
Deep reinforcement learning from human preferences. arXiv
arXiv preprint arXiv:1706.03741, 2017 View on Google Scholar ↗ |
2017 | 51 | 66.1% |
|
Systems and methods for a multi-core optimized recurrent neural network
US Patent 10,832,120, 2020 View on Google Scholar ↗ |
2020 | 29 | 63.3% |
|
Language models are few-shot learners. arXiv [cs. CL]
arXiv preprint arXiv:2005.14165, 2020 View on Google Scholar ↗ |
2020 | 29 | 63.3% |
|
Evaluating large language models trained on code, 2021. doi: 10.48550
arXiv preprint arXiv.2107.03374 10, 2021 View on Google Scholar ↗ |
2021 | 20 | 61.5% |
|
Language models are few-shot learners. arXiv. org
2005 View on Google Scholar ↗ |
2005 | 65 | 61.2% |
|
Scaling laws for neural language models (arXiv: 2001.08361). arXiv
2001 View on Google Scholar ↗ |
2001 | 53 | 56.8% |
|
Low error discrimination using a correlated population code
Journal of neurophysiology 108 (4), 1069-1088, 2012 View on Google Scholar ↗ |
2012 | 36 | 50.2% |
|
Language models are few-shot learners. arXiv [cs. CL]. 2020
arXiv preprint arXiv:2005.14165, 2005 View on Google Scholar ↗ |
2005 | 33 | 47.6% |
|
Better language models and their implications. OpenAI
ed, 2019 View on Google Scholar ↗ |
2019 | 19 | 47.5% |
|
Thermodynamics for a network of neurons: Signatures of criticality
arXiv preprint arXiv:1407.5946, 2014 View on Google Scholar ↗ |
2014 | 26 | 44.5% |
|
Computation of uniform wave forms using complex rays
Physical Review E—Statistical, Nonlinear, and Soft Matter Physics 73 (3 …, 2006 View on Google Scholar ↗ |
2006 | 28 | 44.3% |
|
Language models are few-shot learners.” arXiv [cs. CL]. arXiv
2005 View on Google Scholar ↗ |
2005 | 26 | 43.4% |
|
Language Models Are Few-Shot Learners.” arXiv, July 22, 2020
2005 View on Google Scholar ↗ |
2005 | 25 | 42.7% |
|
Language models are few-shot learners. CoRR, abs/2005.14165
2005 View on Google Scholar ↗ |
2005 | 20 | 39.0% |
|
Microfabricated pipette and method of manufacture
US Patent App. 12/804,442, 2012 View on Google Scholar ↗ |
2012 | 17 | 34.9% |
|
Multi-electrode array recording from salamander retinal ganglion cells
2017 View on Google Scholar ↗ |
2017 | 13 | 33.7% |
|
Language models are few-shot learners. OpenAI
2005 View on Google Scholar ↗ |
2005 | 13 | 32.7% |
|
GPT-2
Language Models are Unsupervised Multitask Learners. OpenAI Blog 1 (9), 2020 View on Google Scholar ↗ |
2020 | 7 | 28.7% |
|
Evaluating Large Language Models Trained on Code.(Jul. 2021)
URL http://arxiv. org/abs/2107.03374, 2021 View on Google Scholar ↗ |
2021 | 5 | 26.8% |
|
Network-scale electrophysiology: measuring and understanding the collective behavior of neural circuits
Princeton University, 2011 View on Google Scholar ↗ |
2011 | 6 | 21.5% |
|
Better language models and their implications
OpenAI Blog https://openai. com/blog/betterlanguage-models, 2019 View on Google Scholar ↗ |
2019 | 4 | 18.8% |
|
Deep speech 2: End-to-end speech recognition in english and mandarin. arXiv
arXiv preprint arXiv:1512.02595, 2016 View on Google Scholar ↗ |
2016 | 5 | 18.6% |
|
Correlations and the structure of the population code in a dense patch of the retina
Proceedings MEA Meeting, 197, 2008 View on Google Scholar ↗ |
2008 | 3 | 14.8% |
|
Searching for collective behavior in a network of real neurons
arXiv preprint arXiv:1306.3061, 2013 View on Google Scholar ↗ |
2013 | 1 | 6.6% |