Layer 4 · Neuron 1743 · suppressed

L4N1743-1

Word quirks:

“⏎”“ ‘”“⏎ ⏎”“ ‘’”“?\”“<|im_end|>”“ --”“.`”

🕒 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)

I am the Chief Information Officer of the University of Notre Dame, and my mission is to ensure that the University's information systems are secure, compliant, and efficient. I will oversee the implementation of policies related to data management, cybersecurity, and IT governance.
</think>

<think>

</think>

<think>

</think>

I am the **Chief Information

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. The Roundabout✅ L2N6196-1	0.97	“ ‘”“⏎”“ ‘’”
2. L4N1107-1🕒	0.96	“⏎”“<\|im_end\|>”“ ‘”
3. L4N122+1🕒	0.95	“<\|im_end\|>”“ ‘”“⏎”
4. L2N7822-1🕒	0.95	“ ‘”“<\|im_end\|>”“⏎”
5. L2N8861+1🕒	0.95	“<\|im_end\|>”“⏎”“ ‘”
6. L2N8878-1🕒	0.95	“<\|im_end\|>”“ ‘”“ ‘’”
7. L5N3491-1💤	0.95	“⏎”“ ‘”“.`”
8. L5N1173-1🕒	0.94	“⏎”“ ‘”“<\|im_end\|>”
9. L4N1133-1🕒	0.94	“<\|im_end\|>”“⏎”“ ‘”
10. L6N2649+1🕒	0.94	“⏎”“ ‘”“ ‘’”

Shadow specimen, apart from the ten: L4N1743+1⛔, -0.19.

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 4, neuron 1743, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.61, top-1 agreement 0.43) 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
L4N1743+1⛔ shadow specimen t0-v0.1	1.045	0.57	1.219	1.52	“ radio”“ honor”“ info”“ mission”“ fire”“ script”“ metaphor”“ reference”
L4N1743-1 t0-v0.1	1.614	0.43	1.887	2.45	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.