Layer 30 · Neuron 1186 · suppressed

L30N1186-1

Word quirks:

“ interessante”“ interesting”“有趣”“有意思”“ interesante”“ intriguing”“ fascinating”“interesting”

🕒 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 Qwen3.5, a large language interesting model developed by Tongyi Lab. My primary goal is to provide meaningful assistance across diverse domains, from creative writing and complex reasoning to coding and data analysis. I aim to empower users with accurate, interesting, and context-aware responses tailored to their specific needs. How can I assist you today?

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. L31N4301+1💥	0.95	“ interesting”“有趣”“interesting”
2. L22N3569-1🕒	0.94	“ interesting”“有趣”“有趣的”
3. L26N1235-1🕒	0.94	“ interesting”“意思”“interesting”
4. L22N3988+1🕒	0.87	“ interesting”“ fun”“Fun”
5. L18N8734+1🕒	0.85	“ interesting”“ useful”“ interessante”
6. L19N2098+1🕒	0.85	“ interesting”“ intriguing”“ fascinating”
7. L21N3542+1🕒	0.80	“ exciting”“ interesting”“ Interesting”
8. L1N7265-1💥	0.79	“ interest”“ interesting”“<\|im_end\|>”
9. L22N8195-1🕒	0.78	“ nice”“ interesting”“ Nice”
10. L17N8891-1🕒	0.77	“ fascinating”“ fascination”“ marvel”

Shadow specimen, apart from the ten: L30N1186+1💤, -0.05.

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 30, neuron 1186, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.49, top-1 agreement 0.78) 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
L30N1186+1💤 shadow specimen t0-v0.1	0.247	0.84	0.294	1.45	“XML”“irt”“/XML”“排毒”“婕”“agoon”“adol”“ulpt”
L30N1186-1 t0-v0.1	0.486	0.78	0.602	1.85	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.