Layer 4 · Neuron 7372 · suppressed

L4N7372-1

Word quirks:

“ ‘”“ ””“<|im_end|>”“ remaining”“ continuation”“ ⏎⏎”“ now”“ ?”

🕒 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 Qwen. My mission is to assist you with any questions about the topic.

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. L11N4437-1🕒	0.75	“ ””“ continuation”“⏎”
2. The Phantom Policy✅ L8N8733-1	0.74	“ ‘”“ ⏎⏎”“<\|im_end\|>”
3. L16N3083-1💤	0.72	“ ‘”“ ⏎”“ ⏎⏎”
4. L12N4873+1💤	0.72	“ now”“<\|im_end\|>”“ ⏎”
5. L11N6048+1🕒	0.68	“ ⏎”“ ‘”“ “”
6. L12N1256+1🕒	0.68	“ ⏎”“ ⏎⏎”“ now”
7. L2N2164+1🕒	0.68	“<\|im_end\|>”“ ””“ ‘’”
8. L10N7742-1💤	0.68	“ ⏎”“<\|im_end\|>”“ ——”
9. L6N693+1💤	0.68	“ ‘”“<\|im_end\|>”“⏎”
10. L8N8304-1🕒	0.68	“<\|im_end\|>”“⏎”“ ‘”

Shadow specimen, apart from the ten: L4N7372+1🕒, 0.07.

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 7372, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.98, top-1 agreement 0.46) 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
L4N7372+1🕒 shadow specimen t0-v0.1	1.093	0.59	1.309	1.48	“ analysis”“ text”“ ‘”“ parse”“ statement”“ textual”“ free”“ analyse”
L4N7372-1 t0-v0.1	1.983	0.46	2.305	2.28	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.