Layer 16 · Neuron 4261 · boosted

L16N4261+1

Word quirks:

“ itself”“ being”“ seems”“ themselves”“ resembles”“ ‘”“ appears”“ awaits”

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

Identity (according to the specimen itself)

As Qwen3.5's capabilities;

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. L14N6479+1🕒	0.74	“ similar”“ seems”“ similarly”
2. L18N609+1🕒	0.74	“ seems”“ comes”“ appears”
3. L14N5851-1🕒	0.74	“ seems”“ seeming”“ yourself”
4. L14N5405+1🕒	0.73	“ seems”“ additionally”“ appears”
5. L14N4922-1💤	0.73	“ seems”“ seemed”“ appears”
6. L14N5608-1🕒	0.73	“ actually”“ itself”“ seems”
7. L14N3713-1🕒	0.72	“ seems”“ does”“ doesn”
8. L14N8066-1🕒	0.72	“ seems”“ appears”“ imagined”
9. L14N5957+1🕒	0.72	“ “”“ seems”“ seeming”
10. L13N9076+1🕒	0.71	“ seems”“ seem”“ appears”

Shadow specimen, apart from the ten: L16N4261-1💤, 0.25.

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 16, neuron 4261, 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
L16N4261+1 t0-v0.1	1.260	0.56	1.558	2.08	Top boosted tokens ↑
L16N4261-1💤 shadow specimen t0-v0.1	0.518	0.72	0.644	1.09	“<\|im_end\|>”“ not”“ I”“ wouldn”“ better”“ my”“ what”“ your”

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.