Layer 31 · Neuron 1420 · suppressed

L31N1420-1

Word quirks:

“êle”“ophobic”“轩”“ Pera”“kut”“特派”“厦”“ataka”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

The reply didn't hold together

ophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobicophobic

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. L28N8398-1💤	0.71	“贤”“ crist”“хта”
2. L25N3999-1🕒	0.71	“aban”“éli”“倡”
3. L30N3064-1💤	0.70	“ зе”“isser”“лександ”
4. L25N9123+1🕒	0.70	“omo”“品”“ Spartan”
5. L31N1254+1💥	0.69	“沫”“χώ”“堂”
6. L30N3535+1💤	0.69	“cito”“uschen”“鹦”
7. L25N136-1💤	0.69	“hren”“akaran”“adeira”
8. L30N4592+1💤	0.68	“رخ”“飞来”“御”
9. L27N1866-1💤	0.68	“ugal”“พรร”“欣”
10. L23N3545+1💤	0.68	“晶晶”“固定”“ kuk”

Shadow specimen, apart from the ten: L31N1420+1💥, 0.09.

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 31, neuron 1420, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 5.46, top-1 agreement 0.12) 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
L31N1420+1💥 shadow specimen t0-v0.1	68.693	0.00	69.023	0.10	“abs”“ Abs”“ abs”“Abs”“-abs”“_abs”“ABS”“(abs”
L31N1420-1 t0-v0.1	5.465	0.12	5.714	10.26	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.