Layer 19 · Neuron 334 · suppressed

L19N334-1

Word quirks:

“ SD”“ SG”“ SVC”“ SA”“ SUS”“ GP”“ CAS”“ PG”

🕒 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 Q, a new model trained on Q(G) data. My mission is to provide high-quality responses in various domains, including but not limited to the following:

Q&A: Answering questions with accuracy and clarity.
DLC: Generating content that meets specific requirements.
Q&A: Solving problems through logical reasoning.
Q&A: Providing information based on training data.
Q&A: Offering insights and knowledge in multiple fields.

If you have any questions or need assistance, I will do my best to provide accurate and relevant answers

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. L14N8956-1🕒	0.81	“ WP”“ CC”“ WS”
2. L30N2699-1🕒	0.80	“ PS”“ SM”“ CS”
3. L22N7776-1🕒	0.77	“ SQ”“ TT”“ SJ”
4. L31N9081-1💥	0.76	“�”“sj”“s”
5. L31N113-1💥	0.75	“ SM”“ ED”“ SC”
6. L27N7540+1🕒	0.74	“SV”“ ES”“ S”
7. L31N2531+1💥	0.74	“sg”“sc”“so”
8. L22N5476-1🕒	0.73	“ SIT”“ TA”“ SES”
9. L30N4447+1🕒	0.73	“S”“G”“SG”
10. L30N1797+1🕒	0.72	“si”“sr”“sd”

Shadow specimen, apart from the ten: L19N334+1💤, 0.05.

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 334, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.62) 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
L19N334+1💤 shadow specimen t0-v0.1	0.472	0.71	0.613	1.11	“ correspondence”“ separator”“ ⏎”“ destination”“ alternative”“ alternativa”“ different”“ifferent”
L19N334-1 t0-v0.1	0.831	0.62	0.982	1.39	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.