Layer 22 · Neuron 2731 · boosted

L22N2731+1

Word quirks:

“ apro”“ WHICH”“ STR”“ WITHOUT”“ pers”“ thereof”“ gian”“ structures”

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

Identity (according to the specimen itself)

The reply didn't hold together

I am QOTER, a large 2.5-100000-300000-80000-5000-0.5-0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000

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. L5N3578+1💥	0.51	“ OF”“LET”“ WHICH”
2. L15N5287-1🕒	0.45	“ which”“ another”“ whose”
3. L27N7183+1🕒	0.44	“amel”“per”“ Per”
4. L7N2812-1🕒	0.44	“for”“wh”“as”
5. L19N6762+1🕒	0.43	“ wherein”“ where”“ whose”
6. L20N1819-1🕒	0.43	“ who”“who”“ pers”
7. L22N238+1🕒	0.43	“stras”“stra”“side”
8. L16N6137+1🕒	0.43	“ of”“ extends”“ that”
9. L16N5512-1🕒	0.42	“ which”“which”“ Which”
10. L18N3446+1💤	0.42	“ of”“ ones”“ for”

Shadow specimen, apart from the ten: L22N2731-1💤, 0.06.

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 22, neuron 2731, 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
L22N2731+1 t0-v0.1	1.151	0.62	1.339	1.80	Top boosted tokens ↑
L22N2731-1💤 shadow specimen t0-v0.1	0.786	0.70	0.754	1.01	“ railway”“Rail”“ Railway”“ crem”“inus”“同样是”“ rail”“ railroad”

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.