Layer 30 · Neuron 5433 · suppressed

L30N5433-1

Word quirks:

“<J”“umon”“ộng”“ljiv”“ haver”“腔”“公墓”“ Jornada”

💤 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 core mission is to become a versatile, capable, and trustworthy AI assistant that can genuinely help users solve problems, create content, and learn—across a broad spectrum of tasks and domains.

Key Aspects of My Mission:

Bridge Human & Machine Intelligence

I aim to bridge the gap between human creativity and computational power, whether through code generation, scientific analysis, or personalized learning support.

Domain-Specialized Support

Whether you need to:

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. L24N6819-1💤	0.67	“oner”“мот”“onge”
2. L25N2599-1💤	0.66	“春风”“usin”“三角”
3. L31N4325+1💥	0.65	“柳”“荆”“荷”
4. L28N4809+1💤	0.65	“귀”“ Ballard”“坚”
5. L30N2218+1💤	0.64	“مل”“ Ses”“鸠”
6. L28N1206+1💤	0.64	“翼”“nehmer”“撞”
7. L30N2250+1💤	0.64	“推移”“迷”“荆”
8. L27N760+1💤	0.63	“驱”“谨”“ Ong”
9. L30N6606-1💤	0.63	“泥”“影”“栏”
10. L23N3405-1💤	0.63	“mong”“anye”“ thereafter”

Shadow specimen, apart from the ten: L30N5433+1💤, 0.34.

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 5433, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.37, top-1 agreement 0.81) 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
L30N5433+1💤 shadow specimen t0-v0.1	0.347	0.86	0.306	1.68	“errated”“alles”“一切”“排在”“ESH”“劫”“ nid”“ORS”
L30N5433-1 t0-v0.1	0.373	0.81	0.557	1.83	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.