Layer 23 · Neuron 5541 · suppressed

L23N5541-1

Word quirks:

“ exhaustion”“豁然”“olla”“خلا”“Cake”“咬牙”“动静”“峻”

💤 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 Qin, a large-scale language model developed by KINGQIN. My primary mission is to provide users with high-quality, accurate, and personalized assistance across a wide range of domains. Here’s how I aim to fulfill this:

Core Capabilities & Goals

Knowledge & Reasoning

Answer questions on topics from science, culture, and technology with factual accuracy.
Solve complex problems through logical reasoning and step-by-step analysis.

Multilingual Support

Communicate in over 100 languages, tailoring responses to cultural

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. L26N8698+1🕒	0.75	“莽”“勃”“odz”
2. L25N3718-1💤	0.75	“เบา”“机能”“ixe”
3. L28N7191-1💤	0.75	“呆”“流”“elos”
4. L25N2347-1💤	0.75	“ON”“绞”“毅然”
5. L31N4441-1💥	0.75	“洞里”“金马”“平地”
6. L27N6516+1💤	0.74	“燃”“曼”“色列”
7. L23N3614-1💤	0.73	“离”“rotch”“λε”
8. L31N5375+1💥	0.73	“得上”“ь”“爽”
9. L28N2823+1💤	0.73	“緩”“urier”“雪”
10. L30N8937-1💤	0.73	“峰”“香山”“峰值”

Shadow specimen, apart from the ten: L23N5541+1🕒, 0.28.

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 5541, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.76) 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
L23N5541+1🕒 shadow specimen t0-v0.1	1.484	0.61	1.612	1.27	“ roaming”“漫游”“ wandering”“ roam”“ wander”“ ro”“游”“-ro”
L23N5541-1 t0-v0.1	0.681	0.76	0.680	1.30	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.