Layer 31 · Neuron 7840 · boosted

L31N7840+1

Word quirks:

“註”“indik”“uchi”“ubar”“oust”“宕”“兜”“trit”

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

Identity (according to the specimen itself)

oustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoustoust

The AI judge wasn't sure this holds together.

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. L31N2809-1💥	0.74	“麻”“寮”“哉”
2. L31N1154+1💥	0.71	“纨”“ubb”“镐”
3. L25N6405-1🕒	0.71	“aper”“乌拉”“肚”
4. L30N4887+1💤	0.71	“umn”“括”“搭”
5. L30N3789+1💤	0.71	“uits”“rut”“tvr”
6. L27N4438+1💤	0.71	“ubar”“仔”“้อน”
7. L29N9170-1🕒	0.70	“承”“�”“ummer”
8. L28N3459+1💤	0.70	“端”“utting”“乌拉”
9. L30N4564-1🕒	0.70	“udz”“omos”“่วง”
10. L30N1689+1💤	0.70	“拨”“冉”“穴”

Shadow specimen, apart from the ten: L31N7840-1💥, 0.44.

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 7840, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 5.61, top-1 agreement 0.09) 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
L31N7840+1 t0-v0.1	5.606	0.09	5.850	9.84	Top boosted tokens ↑
L31N7840-1💥 shadow specimen t0-v0.1	5.519	0.13	5.855	9.90	“zor”“ voelt”“hur”“报名信息”“asurer”“杂”“ริ”“uffed”

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.