| Discovering language model behaviors with model-written evaluations E Perez, S Ringer, K Lukosiute, K Nguyen, E Chen, S Heiner, C Pettit, ... Findings of the Association for Computational Linguistics: ACL 2023, 13387-13434, 2023 | 275 | 2023 |
| Risks from learned optimization in advanced machine learning systems E Hubinger, C van Merwijk, V Mikulik, J Skalse, S Garrabrant arXiv preprint arXiv:1906.01820, 2019 | 170 | 2019 |
| Sleeper agents: Training deceptive llms that persist through safety training E Hubinger, C Denison, J Mu, M Lambert, M Tong, M MacDiarmid, ... arXiv preprint arXiv:2401.05566, 2024 | 150 | 2024 |
| Studying large language model generalization with influence functions R Grosse, J Bae, C Anil, N Elhage, A Tamkin, A Tajdini, B Steiner, D Li, ... arXiv preprint arXiv:2308.03296, 2023 | 139 | 2023 |
| Measuring faithfulness in chain-of-thought reasoning T Lanham, A Chen, A Radhakrishnan, B Steiner, C Denison, ... arXiv preprint arXiv:2307.13702, 2023 | 118 | 2023 |
| Many-shot jailbreaking C Anil, E Durmus, N Panickssery, M Sharma, J Benton, S Kundu, J Batson, ... Advances in Neural Information Processing Systems 37, 129696-129742, 2024 | 109 | 2024 |
| Steering llama 2 via contrastive activation addition N Panickssery, N Gabrieli, J Schulz, M Tong, E Hubinger, AM Turner arXiv preprint arXiv:2312.06681, 2023 | 107 | 2023 |
| Question decomposition improves the faithfulness of model-generated reasoning A Radhakrishnan, K Nguyen, A Chen, C Chen, C Denison, D Hernandez, ... arXiv preprint arXiv:2307.11768, 2023 | 54 | 2023 |
| An overview of 11 proposals for building safe advanced ai E Hubinger arXiv preprint arXiv:2012.07532, 2020 | 30 | 2020 |
| Sycophancy to subterfuge: Investigating reward-tampering in large language models C Denison, M MacDiarmid, F Barez, D Duvenaud, S Kravec, S Marks, ... arXiv preprint arXiv:2406.10162, 2024 | 28 | 2024 |
| Tamera Lanham, Daniel M E Hubinger, JM Carson Denison, M Lambert, M Tong, M MacDiarmid Ziegler, Tim Maxwell, Newton Cheng, et al. Sleeper agents: Training …, 2024 | 28 | 2024 |
| Tamera Lanham, Tim Maxwell, Venkatesa Chandrasekaran, Zac Hatfield-Dodds, Jared Kaplan, Jan Brauner, Samuel R A Radhakrishnan, K Nguyen, A Chen, C Chen, C Denison, D Hernandez, ... Bowman, and Ethan Perez. Question decomposition improves the faithfulness of …, 2023 | 26 | 2023 |
| Alignment faking in large language models R Greenblatt, C Denison, B Wright, F Roger, M MacDiarmid, S Marks, ... arXiv preprint arXiv:2412.14093, 2024 | 22 | 2024 |
| Uncovering deceptive tendencies in language models: A simulated company ai assistant O Järviniemi, E Hubinger arXiv preprint arXiv:2405.01576, 2024 | 17 | 2024 |
| Studying large language model generalization with influence functions, 2023 R Grosse, J Bae, C Anil, N Elhage, A Tamkin, A Tajdini, B Steiner, D Li, ... URL https://arxiv. org/abs/2308.03296, 0 | 15 | |
| Sleeper agents: Training deceptive LLMs that persist through safety training. arXiv E Hubinger, C Denison, J Mu, M Lambert, M Tong, M MacDiarmid, ... | 11 | 2024 |
| Tamera Lanham, Karina Nguyen, Tomasz Korbak, Jared Kaplan, Deep Ganguli, Samuel R. Bowman, Ethan Perez, Roger Grosse, and David Duvenaud. Many-shot jailbreaking C Anil, E Durmus, M Sharma, J Benton, S Kundu, J Batson, N Rimsky, ... Preprint, 2024 | 10 | 2024 |
| AI safety via market making E Hubinger AI Alignment Forum, 2020 | 10 | 2020 |
| Simple probes can catch sleeper agents, 2024 M MacDiarmid, T Maxwell, N Schiefer, J Mu, J Kaplan, D Duvenaud, ... URL https://www. anthropic. com/news/probes-catch-sleeper-agents, 0 | 10 | |
| Engineering monosemanticity in toy models AS Jermyn, N Schiefer, E Hubinger arXiv preprint arXiv:2211.09169, 2022 | 9 | 2022 |
