Layer 19 · Neuron 4693 · suppressed

L19N4693-1

Word quirks:

“ prefix”“ prefixed”“ prefixes”“ naming”“Prefix”“-prefix”“ Prefix”“命名”

🕒 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 Q3, a large model starting with the "Q" prefix. My mission is to provide services that start with "Q" for you:

Query: Answer your questions across various domains.
Question: Solve problems by understanding your needs.
Quality: Deliver high-quality responses tailored to your requirements.
Quickly: Offer efficient and fast interactions.
Quietly: Ensure a smooth experience without technical conflicts.
Questioned: Address any doubts or inquiries you might have.
Questioned: Start with a "Q" to see how we can

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. L31N7905+1💥	0.94	“ prefix”“prefix”“ Prefix”
2. L2N5288+1💥	0.94	“ prefix”“ prefixes”“ Prefix”
3. L25N3628-1🕒	0.94	“ prefix”“ Prefix”“-prefix”
4. L26N2404+1🕒	0.94	“ prefix”“ Prefix”“Prefix”
5. L25N5458-1🕒	0.94	“ prefix”“ Prefix”“Prefix”
6. L20N3405+1🕒	0.94	“ prefix”“ Prefix”“ prefixes”
7. L0N9086-1💥	0.93	“ prefix”“ Prefix”“Prefix”
8. L6N1835+1🕒	0.88	“ prefixes”“ prefix”“ name”
9. L9N4158-1🕒	0.88	“ prefix”“ prefixes”“ prefixed”
10. L19N6284-1🕒	0.87	“ prefixed”“ preceded”“ prefix”

Shadow specimen, apart from the ten: L19N4693+1💤, 0.09.

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 19, neuron 4693, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.00, 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
L19N4693+1💤 shadow specimen t0-v0.1	0.469	0.74	0.592	1.11	“nul”“ null”“ Null”“⏎”“resenter”“egel”“ location”“ tight”
L19N4693-1 t0-v0.1	1.002	0.66	1.064	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.