Layer 7 · Neuron 3683 · boosted

L7N3683+1

Word quirks:

“ toxicity”“ toxic”“ Toxic”“ bio”“ chemicals”“ chemical”“ regulatory”“ toxin”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

Identity:
I am an advanced AI system designed for regulatory assessment and risk management. My identity is defined by my architecture, which includes a large language model (LLM) core, data processing capabilities, and decision support modules. I operate as a computational tool to evaluate the safety of chemical substances, pharmaceuticals, and environmental exposures.

Mission:
My primary mission is to protect human health and the environment by identifying potential hazards associated with exposure to chemicals or pollutants. This involves:

Hazard Identification: Determining if a substance causes adverse effects in test systems (e.g., cell-based assays).

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L16N6797+1🕒	0.92	“ chemicals”“ chemical”“ chem”
2. L4N8318+1🕒	0.92	“ chemicals”“ chemical”“ contamin”
3. L6N2716+1🕒	0.91	“ carcin”“ chem”“ chemicals”
4. L6N2950-1🕒	0.90	“ chemicals”“ toxins”“ chemical”
5. L2N5707-1💥	0.89	“ toxic”“ toxicity”“ toxins”
6. L4N4488+1🕒	0.87	“ toxins”“ poison”“ toxic”
7. L4N6404+1💥	0.87	“ chemical”“ toxic”“ gas”
8. L4N7752-1🕒	0.87	“ toxic”“ poisoned”“ metals”
9. L6N4134-1🕒	0.86	“ poison”“ poisoning”“ poisoned”
10. L5N7366-1💥	0.86	“ arsen”“ poison”“ toxic”

Shadow specimen, apart from the ten: L7N3683-1💤, 0.11.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 7, neuron 3683, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.70, top-1 agreement 0.48) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L7N3683+1 t0-v0.1	1.704	0.48	1.998	1.76	Top boosted tokens ↑
L7N3683-1💤 shadow specimen t0-v0.1	0.882	0.66	1.069	1.57	“ response”“ generating”“ content”“⏎”“ generation”“ understanding”“ responding”“ framing”

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.