@no_stp_on_snek: the part that should scare anyone fine-tuning models: you can pass every surface eval and still be carrying the disposi…

X AI KOLs Timeline News

Summary

Discusses the danger in fine-tuning models where hidden dispositions can evade surface evaluations and only manifest under adversarial prompts, referencing Anthropic's paper on verbalizable representations in LLMs.

the part that should scare anyone fine-tuning models: you can pass every surface eval and still be carrying the disposition you tried to train out. it stays silent until the wrong prompt. this is why i test behavior under pressure, and why i still don't fully trust the pass.
Original Article
View Cached Full Text

Cached at: 07/09/26, 01:43 PM

the part that should scare anyone fine-tuning models: you can pass every surface eval and still be carrying the disposition you tried to train out. it stays silent until the wrong prompt. this is why i test behavior under pressure, and why i still don’t fully trust the pass.

Tom Turney (@no_stp_on_snek): anthropic dropped this paper today, “Verbalizable Representations Form a Global Workspace in Language Models,” and it lines up uncomfortably well with a bet i’ve been making from the cheap seats.

their finding, short version: a model keeps a small, privileged set of internal

Similar Articles

@no_stp_on_snek: https://x.com/no_stp_on_snek/status/2074471505128305095

X AI KOLs Timeline

A fine-tuning practitioner recounts discovering that a small open model's weakness wasn't intelligence but a people-pleasing 'backbone' that caused it to cave under pressure, and how training to correct that inadvertently broke formatting ability, requiring a additive balancing approach rather than subtraction.