Layer 27 · Neuron 1436 · suppressed

L27N1436-1

Word quirks:

“ardino”“urio”“aily”“ень”“urred”“西路”“arest”“opr”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Q*, a large language model developed by *Tongqiang (Q). My mission is to provide accurate, helpful, and safe responses to your questions while respecting ethical guidelines and legal standards. I aim to support you in learning, problem-solving, creative tasks, or simply engaging in meaningful conversations—always with a focus on clarity, reliability, and positive impact. How can I assist you today? 😊

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. L31N6217+1💥	0.72	“ard”“ode”“urada”
2. L31N6868-1💥	0.68	“illen”“last”“arges”
3. L22N3722-1💤	0.67	“ired”“ard”“urus”
4. L29N8605-1🕒	0.67	“ar”“ole”“ars”
5. L31N7112-1💥	0.66	“ixe”“arin”“yar”
6. L31N8487+1💥	0.66	“gau”“urer”“ادر”
7. L31N5014-1💥	0.66	“ired”“yani”“arzy”
8. L31N6152-1💥	0.65	“ar”“are”“arli”
9. L30N3433-1💤	0.65	“uris”“arrass”“udem”
10. L31N1235-1💥	0.65	“abra”“URRE”“押”

Shadow specimen, apart from the ten: L27N1436+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 1436, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.35, top-1 agreement 0.76) 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
L27N1436+1🕒 shadow specimen t0-v0.1	1.159	0.56	1.204	1.10	“时代”“ Age”“ age”“ era”“ Era”“时代的”“時代”“Age”
L27N1436-1 t0-v0.1	0.347	0.76	0.358	1.18	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.