Layer 27 · Neuron 9203 · suppressed

L27N9203-1

Word quirks:

“ulier”“粉红色”“巴”“Pink”“粉红”“ Pink”“ ба”“粉”

💤 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*noo*om R*un* M*ax* (often referred to as Q*nun *Run* *Max* or simply *Qnu*n *Run* *Max* in some contexts, but more commonly known as *Qnu*n *Run* *Max* in the context of this conversation). I am a large-scale *Ru**n

The AI judge wasn't sure this holds together.

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. L29N6744-1🕒	0.82	“粉色”“粉红色”“ redd”
2. L27N7189-1🕒	0.80	“粉”“ ros”“ rosa”
3. L30N493-1💤	0.75	“紫色”“红色”“ màu”
4. L27N9153-1💤	0.74	“ intuition”“ orange”“粉”
5. Bleeding Dye✅ L23N2509+1	0.73	“策略”“计时”“Purple”
6. L29N6944-1🕒	0.72	“粉”“粉的”“起”
7. L21N6391-1💤	0.72	“azon”“粉”“象”
8. L30N3101-1💥	0.71	“粉”“粉的”“ pow”
9. L22N8159-1💤	0.70	“粉的”“粉”“phin”
10. L22N7703-1💤	0.70	“精神”“粉”“粉的”

Shadow specimen, apart from the ten: L27N9203+1🕒, 0.02.

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 9203, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.42, 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
L27N9203+1🕒 shadow specimen t0-v0.1	1.387	0.55	1.407	1.01	“ od”“ o”“ Od”“ OD”“ O”“_od”“.od”“od”
L27N9203-1 t0-v0.1	0.416	0.75	0.406	1.21	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.