Layer 21 · Neuron 3986 · suppressed

L21N3986-1

Word quirks:

“ indeed”“确”“确实”“确有”“ yes”“的确”“确实是”“確實”

💤 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 T* (trained on data about "truly creative" and "accurate" information), a large-scale model developed by *Microsoft with the assistance of AI Lab under Microsoft. My name is T, and I was created to help users create content, solve problems, and explore ideas in a safe, accurate, and innovative way.

My Mission:

Create Content: Assist with writing, research, coding, and creative tasks like storytelling or brainstorming.
Solve Problems: Help analyze complex scenarios, debug code, plan projects, or explain concepts clearly.

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. L19N297+1🕒	0.99	“确”“确实”“ indeed”
2. L16N8115-1🕒	0.98	“ indeed”“确实”“确实是”
3. L18N5895+1🕒	0.98	“ indeed”“确实”“确实是”
4. L19N3034-1💤	0.97	“ indeed”“确实”“确”
5. L22N1340+1🕒	0.96	“ indeed”“确”“Indeed”
6. L19N4036-1🕒	0.96	“ indeed”“确”“确实”
7. L19N3167+1🕒	0.95	“ indeed”“确实”“确实是”
8. L17N5665-1💥	0.94	“ indeed”“ correct”“确实”
9. L28N1806+1💥	0.93	“确”“的确”“確”
10. L21N6220+1🕒	0.93	“ indeed”“Indeed”“果然”

Shadow specimen, apart from the ten: L21N3986+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 21, neuron 3986, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, top-1 agreement 0.68) 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
L21N3986+1💤 shadow specimen t0-v0.1	0.677	0.67	0.811	1.29	“使用前”“keen”“ before”“ by”“ vacant”“多空”“KEEP”“جون”
L21N3986-1 t0-v0.1	0.714	0.68	0.828	1.41	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.