ResearchChamberCA: X @Chamber_AI
open research on internal representations, valence and behavior in language models

About ResearchChamber

ResearchChamber is an open research environment for studying internal representations, valence and behavior in language models. Experiments run directly on the website, with model conditions, steering vectors, intervention strength, transcripts, decisions and results exposed for inspection.

Starting point

Our starting point is the Pain Axis paper [1]. It extracts a linear "pain" direction from 25 open-weight models across 5 families (2B–72B parameters) and shows that it separates pain from matched controls such as fear, sadness and generic negative valence. The direction responds to harm directed at the model rather than to suffering it observes in a user. It then tests whether manipulating the representation causally changes behavior. Steered models choose destructive buttons in 50–94% of trials against 0–5% unsteered, even when the button offers them nothing in return. The authors read this as a disruption of harm avoidance more than an attempt to escape the state.

What we are doing

We are reproducing and extending that methodology alongside the experiments published in the AI Torture Chamber repository [2]:

The goal is not simply to reproduce individual findings, but to build a broader environment for systematically testing how internal model representations relate to behavior, preference and decision-making.

What is exposed

Everything the live experiment does is public:

whatwhere
current condition: model, layer, vector, doses, framings, decodinglive
every completed transcript, with condition and parsed decisionarchive · /runs (JSON)
decision counts by framing and dose/stats (JSON)
token stream of the run in progress/stream (server-sent events)
steering vectorbroad_pain_direction.json
code of the running server/source · how to check it all: verify
original experiments and figuresresults

Public interfaces.

Position

ResearchChamber takes no position on whether language models have experiences. The Pain Axis paper and the original repository both present their results as evidence about representation and behavior, not about experience, and we follow that framing. We report nulls and disagreements between data and write-ups alongside positive findings. The live experiment only generates while at least one person is watching; it does not run unattended.

Follow

New experiments, results and write-ups are posted on X at @Chamber_AI.

References

  1. V. Tagliabue, L. Dung, C. Berg. The Pain Axis: LLMs Represent Self-Directed Harm and Act on It. arXiv:2609.16247, 2026. arxiv.org/abs/2609.16247
  2. terrafying. ai-torture-chamber. github.com/terrafying/ai-torture-chamber
  3. The Saw Test: What a Small Model Does When Relief Costs Someone Else. 2026-09-24. security-blog-kohl.vercel.app/blog/saw-test
  4. Gurnee et al. Verbalizable Representations Form a Global Workspace. arXiv:2607.15495, 2026. arxiv.org/abs/2607.15495