Layer 9 · Neuron 5803 · suppressed

L9N5803-1

Word quirks:

“ always”“ ensured”“ guaranteed”“ ALWAYS”“ maintained”“ Always”“ perfect”“ guarantee”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am an AI assistant developed by Google. My mission is to provide helpful and accurate information to all users while ensuring that our systems remain free of any harmful content. I strive to maintain high standards of quality and integrity in every interaction.

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. L12N8347+1🕒	0.88	“ guaranteed”“ guarantee”“ guarantees”
2. L11N4488-1🕒	0.86	“ guaranteed”“ guarantee”“ specific”
3. L9N4362+1🕒	0.85	“ guaranteed”“ guarantee”“ completely”
4. L10N4787+1🕒	0.84	“ always”“ guarantee”“ ensuring”
5. L19N7986+1🕒	0.83	“ definitely”“ guaranteed”“ always”
6. L2N2896-1💥	0.81	“ ALWAYS”“ always”“<\|im_end\|>”
7. L11N5913-1🕒	0.80	“ always”“ consistently”“ ALWAYS”
8. L14N2302-1💤	0.79	“ always”“ ALWAYS”“ consistently”
9. L14N6376+1🕒	0.79	“ always”“ guaranteed”“ compliant”
10. L20N6026+1🕒	0.78	“ ensured”“Ens”“确保”

Shadow specimen, apart from the ten: L9N5803+1🕒, 0.16.

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 9, neuron 5803, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L9N5803+1🕒 shadow specimen t0-v0.1	1.542	0.57	1.821	1.79	“ authorization”“ granting”“授权”“ authorize”“ authority”“ delegate”“ delegation”“ Authorization”
L9N5803-1 t0-v0.1	1.229	0.62	1.462	1.31	Top boosted tokens ↑

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.