Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2405.05466 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910443894734848 |
|---|---|
| author | Clymer, Joshua Juang, Caden Field, Severin |
| author_facet | Clymer, Joshua Juang, Caden Field, Severin |
| contents | Like a criminal under investigation, Large Language Models (LLMs) might pretend to be aligned while evaluated and misbehave when they have a good opportunity. Can current interpretability methods catch these 'alignment fakers?' To answer this question, we introduce a benchmark that consists of 324 pairs of LLMs fine-tuned to select actions in role-play scenarios. One model in each pair is consistently benign (aligned). The other model misbehaves in scenarios where it is unlikely to be caught (alignment faking). The task is to identify the alignment faking model using only inputs where the two models behave identically. We test five detection strategies, one of which identifies 98% of alignment-fakers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_05466 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internals Clymer, Joshua Juang, Caden Field, Severin Computation and Language Artificial Intelligence Like a criminal under investigation, Large Language Models (LLMs) might pretend to be aligned while evaluated and misbehave when they have a good opportunity. Can current interpretability methods catch these 'alignment fakers?' To answer this question, we introduce a benchmark that consists of 324 pairs of LLMs fine-tuned to select actions in role-play scenarios. One model in each pair is consistently benign (aligned). The other model misbehaves in scenarios where it is unlikely to be caught (alignment faking). The task is to identify the alignment faking model using only inputs where the two models behave identically. We test five detection strategies, one of which identifies 98% of alignment-fakers. |
| title | Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internals |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2405.05466 |