Validation-Gated Causal Interventions for Interpreting High-Stakes Large Language Model Behavior: A Case Study in Suicidality Detection

Abstract:Large language models are proposed to flag suicidal content, but representing a concept and acting on it are distinct. A suicidality feature is decodable from 0.5B parameters upward, while the behavior that uses it emerges only in the low billions. Where it emerges, it rests on a compact mid-network feature: in Llama-3.1-8B-Instruct this direction decodes suicidal from non-suicidal text above a matched lexical baseline and survives removal of the 38 most label-predictive keywords. Projecting it out drops the model's ranking from AUC 0.975 to 0.832, while an equal-norm random direction does not; ablating a low-dimensional subspace degrades it further, though no stable rank is recoverable. The direction transfers to two further corpora, replicates in Qwen and Mistral, and tracks clinician-assigned risk. Steering raises an off-target readout as much as the target across five concepts, so sufficiency claims from yes/no readouts are confounded by generic salience.

Computation and Language (cs.CL)

arXiv:2606.21078 [cs.CL]

(or arXiv:2606.21078v2 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2606.21078 Focus to learn more arXiv-issued DOI via DataCite

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.