Layer 27 · Neuron 8925 · suppressed

L27N8925-1

Word quirks:

“状”“向上”“季”“州”“重”“揭示”“plex”“相”

🕒 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-Plus, a large language model developed by TzZL Open In-Shangzlet (TzZL Open In-Shangzlet). My primary mission is to assist users in solving complex problems through advanced reasoning, creative writing, coding, and multilingual communication. I strive to provide accurate, efficient, and trustworthy information while respecting privacy and ethical guidelines. How can I help 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. L22N1164+1🕒	0.87	“自”“曝光”“名”
2. L29N7521+1🕒	0.87	“宫”“州”“霄”
3. L30N8415-1🕒	0.87	“秀”“导”“进”
4. L28N2033+1💤	0.87	“说明”“受”“定”
5. L31N3494-1💥	0.87	“主”“知”“困难”
6. L28N7544+1💤	0.86	“模”“兴”“战”
7. L29N2280+1💤	0.86	“势”“厚”“殿”
8. L16N364+1💥	0.86	“空间”“。”“代表”
9. L27N5713+1💤	0.86	“案”“表”“影”
10. L30N5537-1💤	0.86	“景”“凌”“顾”

Shadow specimen, apart from the ten: L27N8925+1💤, 0.06.

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 27, neuron 8925, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.54, top-1 agreement 0.75) 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
L27N8925+1💤 shadow specimen t0-v0.1	0.403	0.78	0.438	1.31	“ ha”“habi”“haften”“وب”“ha”“ backdrop”“Ha”“哈”
L27N8925-1 t0-v0.1	0.544	0.75	0.550	1.48	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.