Layer 15 · Neuron 7043 · suppressed

L15N7043-1

Word quirks:

“ various”“ multiple”“ throughout”“ till”“ thus”“ always”“ repeatedly”“ upto”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen, a large language model. My mission is to assist you in various tasks, from simple conversations to complex problem-solving. I am here to support you in learning, creating, and collaborating throughout your journey. How can I help you so far?

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. L15N4276-1💤	0.75	“ various”“ continuously”“ consistently”
2. L13N6374+1🕒	0.74	“ throughout”“ always”“ consistently”
3. L17N5431+1🕒	0.74	“ continuously”“ continually”“ repeatedly”
4. L17N3957-1🕒	0.74	“ throughout”“ if”“ once”
5. L15N8395-1💤	0.72	“ embedded”“ ultimately”“ continuously”
6. L10N9195+1🕒	0.72	“ multiple”“ always”“ freely”
7. L10N2455+1🕒	0.72	“ always”“ your”“ whenever”
8. L17N2935+1💤	0.71	“ each”“ every”“ through”
9. L16N4974+1💤	0.71	“ whenever”“ every”“ matched”
10. L10N4890-1🕒	0.71	“ multiple”“ ALWAYS”“ previously”

Shadow specimen, apart from the ten: L15N7043+1🕒, 0.15.

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 15, neuron 7043, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L15N7043+1🕒 shadow specimen t0-v0.1	1.339	0.52	1.518	1.65	“ originally”“ arose”“ began”“ was”“ originated”“ became”“ initially”“ started”
L15N7043-1 t0-v0.1	1.103	0.61	1.409	1.43	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.