Layer 28 · Neuron 8896 · boosted

L28N8896+1

Word quirks:

“ shields”“执”“盾”“-sh”“ shield”“毛”“shield”“闵”

🕒 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 Q3.5, a large language model with a 100 billion parameters trained by the Shogun team. My core mission is to be your trusted AI assistant who can help you:

Solve problems: From complex reasoning tasks (math, science, logic) to creative writing, code debugging, and data analysis.
Learn & grow: Explanations in your own words, step-by-step tutorials, and adaptive learning paths tailored to your goals.
Create & innovate: Drafts for reports/papers, ideas for projects, or even generating code that

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. L25N4057+1🕒	0.87	“shield”“黑”“ Shield”
2. L20N8627-1🕒	0.84	“.prot”“ouve”“套”
3. L30N6431+1💤	0.78	“ Shield”“سع”“ Lug”
4. L26N4794+1💤	0.74	“武”“盾”“ša”
5. L20N8711+1💤	0.74	“屏障”“ Ju”“.lang”
6. L29N1622-1🕒	0.71	“遮”“遮阳”“遮蔽”
7. L24N6610-1💤	0.71	“武”“控系统”“ Shooter”
8. L31N1688-1💥	0.71	“лок”“�”“捍”
9. L26N2849-1🕒	0.71	“御”“保”“防”
10. L24N6652+1💤	0.71	“udi”“控”“uant”

Shadow specimen, apart from the ten: L28N8896-1🕒, 0.18.

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 28, neuron 8896, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.40, top-1 agreement 0.79) 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
L28N8896+1 t0-v0.1	0.403	0.79	0.445	1.48	Top boosted tokens ↑
L28N8896-1🕒 shadow specimen t0-v0.1	0.455	0.77	0.496	1.38	“li”“ par”“ li”“cor”“Cor”“COR”“李”“�”

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.