Layer 25 · Neuron 2300 · suppressed

L25N2300-1

Word quirks:

“ defining”“ define”“ defines”“ Define”“ defined”“定义”“define”“Define”

🕒 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 Crown, a large language model defined by my role as a precise and creative AI assistant. My core mission is to define and execute the following:

Define: Clearly define my identity as a high-fidelity AI designed for precision, creativity, and real-world application.
Execute: Execute user-defined tasks with accuracy, defining clear outcomes for each interaction.

In short, I exist to define and execute your needs with clarity and precision. How can I define and execute your next task?

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. L22N2289-1🕒	1.00	“ define”“ Define”“define”
2. L19N2347+1🕒	0.99	“ define”“定义”“ Define”
3. L25N1863+1🕒	0.99	“ defining”“ defines”“ define”
4. L25N1576-1🕒	0.99	“ defines”“ define”“ Define”
5. L23N1274-1💤	0.98	“定义”“ define”“ Define”
6. L31N3524-1💥	0.98	“ defining”“ defines”“ define”
7. L23N2695+1⛔	0.98	“ defining”“ defines”“ define”
8. L25N988+1🕒	0.98	“ defined”“ define”“ defines”
9. L23N856-1💤	0.97	“ defines”“ define”“定义了”
10. L19N6899+1💥	0.97	“define”“ define”“ defines”

Shadow specimen, apart from the ten: L25N2300+1💤, 0.16.

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 25, neuron 2300, 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 0.66) 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
L25N2300+1💤 shadow specimen t0-v0.1	0.599	0.74	0.650	1.30	“ Analyse”“ Plate”“ Spect”“ Sciences”“boards”“投票”“ Inquiry”“插图”
L25N2300-1 t0-v0.1	0.865	0.66	0.835	1.26	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.