Layer 23 · Neuron 8240 · suppressed

L23N8240-1

Word quirks:

“主人”“ Owners”“.owner”“ owners”“owner”“ Owner”“管理层”“ owner”

🕒 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 Q3D, a high-quality text version of the 3D version of the Q3D series. I am designed to be a versatile and intelligent assistant capable of handling a wide range of tasks with high accuracy and efficiency.

My mission is to serve as a reliable partner in your journey of exploration and creation. Whether you need to:

Explore complex information across different domains (like science, culture, or technology),
Create high-quality content such as stories, scripts, or code,
Manage complex tasks like planning projects or analyzing data,

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. L22N4758+1🕒	0.97	“ owners”“ Owners”“ Owner”
2. L21N8395-1🕒	0.96	“ owner”“ owners”“主人”
3. L29N7437-1🕒	0.96	“owner”“ Owner”“ owner”
4. L23N7113-1🕒	0.96	“Owner”“ owners”“ Owners”
5. L22N4461-1🕒	0.95	“ owner”“ Owner”“owner”
6. L22N7336-1💤	0.95	“主人”“ owner”“_owner”
7. L23N7599+1💤	0.95	“业主”“ Owners”“ owners”
8. L21N8863-1🕒	0.93	“-owner”“ Owner”“ owner”
9. L22N6884+1🕒	0.93	“ owners”“车主”“ owner”
10. L20N4922-1💤	0.92	“ owners”“ owner”“owner”

Shadow specimen, apart from the ten: L23N8240+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 8240, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, 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
L23N8240+1🕒 shadow specimen t0-v0.1	1.074	0.64	1.107	1.37	“ meeting”“meeting”“Meeting”“ Meeting”“ session”“ meets”“ meet”“ sess”
L23N8240-1 t0-v0.1	0.859	0.68	0.901	1.22	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.