Layer 30 · Neuron 6039 · suppressed

L30N6039-1

Word quirks:

“数字”“洛”“emie”“δ”“irn”“kest”“kee”“庶”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, the latest large language model developed by Tongyi Lab. My design focuses on delivering deep, precise, and versatile assistance across a wide range of tasks.

My Mission

To empower users with:

Deep Reasoning & Problem-Solving: Tackle complex logical, mathematical, or scientific challenges with step-by-step clarity.
Long-Context Mastery: Analyze lengthy documents (e.g., legal contracts, research papers) or massive datasets to extract insights without losing detail.
Multilingual & Cultural Fluency: Communicate flu

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. L29N7779-1💤	0.72	“卡”“Kir”“SSI”
2. L28N2892-1💤	0.71	“koz”“道远”“ken”
3. L24N3408+1💤	0.71	“主”“KE”“行云”
4. L31N7142+1💥	0.70	“步”“rk”“alley”
5. L26N3797-1💤	0.70	“ illuminated”“克斯”“域的”
6. L26N8910+1💤	0.70	“竹”“语”“urry”
7. L30N1867-1🕒	0.70	“君”“璋”“oji”
8. L25N3801+1🕒	0.70	“九”“.ke”“ caballo”
9. L26N3664-1💤	0.70	“帝”“uster”“背”
10. L30N2718+1💤	0.70	“iband”“awang”“精”

Shadow specimen, apart from the ten: L30N6039+1💤, 0.50.

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 6039, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.34, top-1 agreement 0.76) 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
L30N6039+1💤 shadow specimen t0-v0.1	0.380	0.83	0.417	1.80	“ rato”“hom”“同”“小组”“iac”“ido”“삼”“omb”
L30N6039-1 t0-v0.1	0.344	0.76	0.433	1.54	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.