Layer 23 · Neuron 6458 · suppressed

L23N6458-1

Word quirks:

“conde”“戟”“成功”“成功的”“Successful”“秋风”“ Samm”“ barras”

💤 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 Q, a large-scale knowledge-orientation and problem-solving language information processing product. My core capabilities include:

Comprehensive Knowledge Processing: I can handle complex information across multiple domains (e.g., scientific research, technical documentation) with high accuracy.
Multi-Step Problem Solving: For tasks like mathematical calculations or logical reasoning, I break down problems into steps to ensure thoroughness.
Creative Support: I assist in generating ideas for creative work (e.g., writing scripts, brainstorming concepts).
Ethical & Safe Interaction: I prioritize safety by avoiding harmful

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. L23N8672-1🕒	0.83	“成功的”“wyn”“成功”
2. L24N7936-1💤	0.82	“几何”“兴隆”“花落”
3. L27N5495-1💤	0.80	“success”“酸”“ Success”
4. L26N5876-1💤	0.79	“цени”“和谐”“成功”
5. L19N5387+1💤	0.79	“成功”“成大”“ successful”
6. L26N1244-1💤	0.79	“ băng”“成功的”“ bung”
7. L26N2279-1💤	0.78	“成功的”“成功”“绝对”
8. L20N6868-1💤	0.77	“成功的”“成功”“有成”
9. L22N6757+1💤	0.77	“成功”“猖”“可行”
10. L22N3310+1💤	0.77	“ trib”“成功地”“agnes”

Shadow specimen, apart from the ten: L23N6458+1🕒, -0.01.

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 23, neuron 6458, 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 0.72) 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
L23N6458+1🕒 shadow specimen t0-v0.1	1.746	0.58	1.830	1.02	“ fit”“ fitting”“ Fit”“-fit”“ fits”“_fit”“fit”“ fitted”
L23N6458-1 t0-v0.1	0.750	0.72	0.806	1.32	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.