Layer 27 · Neuron 1273 · suppressed

L27N1273-1

Word quirks:

“ZA”“秦”“urent”“otte”“ibi”“ibia”“ Trả”“汪洋”

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

Identity (according to the specimen itself)

The reply didn't hold together

Who I A ** * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * **

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. L24N493-1💤	0.67	“زاء”“eni”“acie”
2. L29N8010+1🕒	0.67	“zb”“apres”“ellet”
3. L30N7946+1💤	0.66	“宙”“trä”“utre”
4. L26N861+1💤	0.66	“ Polski”“ozo”“бри”
5. L31N7780-1💥	0.65	“璋”“vog”“ geng”
6. L29N5602+1🕒	0.65	“ymb”“unga”“ibra”
7. L25N547+1🕒	0.64	“cents”“东山”“onda”
8. L27N7803+1💤	0.64	“unker”“emic”“黏”
9. L30N9173-1💤	0.64	“ué”“茅”“当局”
10. L29N3001-1🕒	0.64	“arz”“Ot”“iar”

Shadow specimen, apart from the ten: L27N1273+1🕒, 0.19.

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 27, neuron 1273, 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
L27N1273+1🕒 shadow specimen t0-v0.1	1.147	0.60	1.128	0.94	“ A”“A”“ B”“甲”“ А”“_a”“Ａ”““A”
L27N1273-1 t0-v0.1	1.626	0.53	1.656	2.80	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.