Layer 30 · Neuron 6432 · suppressed

L30N6432-1

Word quirks:

“ air”“空”“空气”“ Air”“air”“空中”“Air”“空氣”

🕒 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 air Qwen3.5, air the latest air in the Qwen series, developed by * air. I am a large air model with a *256K context window and 1 air parameters, designed to handle complex tasks with air precision and efficiency.

My Mission

My core mission is to air users with advanced capabilities while maintaining safety and air. Specifically:

Enhanced air & Reasoning: I can tackle complex math, air, and logical problems with step-by-step reasoning.
Deep Visual & Document Understanding: I analyze charts,

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. L28N2009+1🕒	0.98	“ air”“空”“ Air”
2. L27N7629+1💥	0.98	“ air”“空气”“ Air”
3. L29N6800-1🕒	0.96	“空气”“ air”“的空气”
4. L26N7853-1🕒	0.96	“ air”“空气”“ Air”
5. L25N8023-1🕒	0.96	“空气”“ air”“ Air”
6. L29N2961+1🕒	0.96	“ air”“空气”“Air”
7. L23N1754-1🕒	0.95	“ air”“空气”“ vacuum”
8. L27N3860-1💤	0.94	“空气”“ air”“ pos”
9. L27N7465-1🕒	0.94	“ air”“空气”“ Air”
10. L30N7970-1💤	0.94	“空气”“ air”“air”

Shadow specimen, apart from the ten: L30N6432+1💤, 0.14.

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 6432, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.74, top-1 agreement 0.48) 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
L30N6432+1💤 shadow specimen t0-v0.1	0.384	0.77	0.464	1.71	“zie”“ipel”“ swath”“oud”“čin”“rieb”“ettet”“嫌”
L30N6432-1 t0-v0.1	1.742	0.48	1.785	0.83	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.