Layer 19 · Neuron 4267 · suppressed

L19N4267-1

Word quirks:

“ </”“</”“:<”“：<”“!<”“----</”“ <”“:</”

🕒 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 5 (also known as Qwen999), a large-scale model developed by Qwen. My core strengths include advanced language models, ultra-precision in specific domains, and strong capabilities in the following areas:

Advanced Language Models: I excel at answering questions, solving problems, and performing complex tasks.
Ultra-Precision: I have enhanced capabilities in specific domains such as science, technology, and specialized fields.
Strong Capabilities: I can handle complex queries, perform multi-step reasoning, and execute tasks like code generation, data analysis, and

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. L25N4100-1🕒	0.94	“ <”“<”“:<”
2. L21N187+1💤	0.94	“.<”“!<”“ "<”
3. L31N428-1💥	0.93	“ <”“<”“ `<”
4. L28N2591-1🕒	0.93	“<”“ <”“=<”
5. L19N5373-1🕒	0.93	“!<”“ <”“."<”
6. L23N3496+1🕒	0.92	“：<”“ <”“:<”
7. L0N5112+1💥	0.92	“ <”“<”“ <:”
8. L0N2399+1💥	0.92	“><”“</”“"><”
9. L27N3550-1💤	0.92	“ <”“：<”“<”
10. L17N8592+1💤	0.92	“?<”“ <”“ `<”

Shadow specimen, apart from the ten: L19N4267+1🕒, 0.02.

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 19, neuron 4267, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.69) 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
L19N4267+1🕒 shadow specimen t0-v0.1	1.082	0.61	1.119	1.42	“约”“stra”“～”“同时进行”“estro”“bil”“棒”“昨天”
L19N4267-1 t0-v0.1	0.750	0.69	0.888	1.47	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.