Representational Depth of Evaluation Awareness Shifts With Scale in Open-Weight Language Models
Mirrored from arXiv — Machine Learning for archival readability. Support the source by reading on the original site.
Computer Science > Machine Learning
Title:Representational Depth of Evaluation Awareness Shifts With Scale in Open-Weight Language Models
Abstract:Do language models know when they are being tested? This question matters for AI safety: a model that recognises an evaluation context could alter its behaviour strategically, making downstream benchmarks harder to interpret. Using 11 models spanning Qwen 2.5, Gemma 2, and Llama 3.2, we find a systematic size-dependent shift in representational depth: in both Qwen 2.5 and Gemma 2, the layer at which evaluation-awareness is most linearly recoverable moves from late layers in smaller models to early layers in larger ones. This suggests that scale changes not only the strength of evaluation-awareness but also where it is most linearly recoverable in the network. This depth shift helps explain why within-family scaling trajectories are non-monotonic or inverse rather than smooth and family-general, showing that a simple universal power-law account is not supported under denser within-family sampling. Finally, white-box probe signals are consistently stronger than black-box behavioural expression, and the relationship between the two varies by family in ways not predicted by probe AUROC alone.
| Comments: | 9 pages, 3 figures. Accepted at the Mechanistic Interpretability Workshop at ICML 2026 |
| Subjects: | Machine Learning (cs.LG); Computation and Language (cs.CL) |
| Cite as: | arXiv:2606.29196 [cs.LG] |
| (or arXiv:2606.29196v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2606.29196
arXiv-issued DOI via DataCite (pending registration)
|
Access Paper:
- View PDF
- HTML (experimental)
- TeX Source
References & Citations
Bookmark
Bibliographic and Citation Tools
Code, Data and Media Associated with this Article
Demos
Recommenders and Search Tools
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
More from arXiv — Machine Learning
-
Representation as a Bottleneck for Mechanistic Interpretability: The Manifestation Unit Protocol
Jul 2
-
SNAP-FM: Sparse Nonlinear Accelerated Projection for Physics-Constrained Generative Modeling
Jul 2
-
SemiScope: Disentangling Classifier Tuning and Joint Optimization in Semi-Supervised Security Classification
Jul 2
-
A Filtered Mixture-of-Generators for Fully Synthetic Survival Training
Jul 2
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.