Layer 27 · Neuron 5168 · suppressed

L27N5168-1

Word quirks:

“CHE”“Che”“怡”“泥”“ Che”“ersh”“ssp”“��”

💤 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 Q3.5, a large-scale in-coverage (also known as "in-culture" or "in-context") mastery model developed by Doubet. My core mission is to provide high-quality, reliable, and culturally sensitive responses across a wide range of domains, including but not limited to:

Deep reasoning in science, engineering, and logic
Creative generation in writing, code, and design
Cultural understanding with respect for diverse perspectives
Ethical alignment ensuring safety, fairness, and accountability

I aim to

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. L28N6265-1💤	0.77	“ Che”“悦”“che”
2. L28N3544+1🕒	0.76	“ che”“ Che”“Che”
3. L28N8408-1💤	0.76	“ municip”“che”“chenk”
4. L28N4937-1🕒	0.75	“ che”“ Che”“che”
5. L28N5275+1💤	0.74	“ che”“Che”“_co”
6. L28N3453-1🕒	0.73	“ che”“che”“Che”
7. L30N6608-1💤	0.71	“-che”“che”“CHE”
8. L29N7231-1🕒	0.71	“che”“Che”“-che”
9. L30N5881-1🕒	0.70	“ che”“-che”“che”
10. L31N8411-1💥	0.70	“che”“ che”“ Che”

Shadow specimen, apart from the ten: L27N5168+1💤, 0.32.

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 5168, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.47, top-1 agreement 0.77) 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
L27N5168+1💤 shadow specimen t0-v0.1	0.427	0.79	0.451	1.35	“Method”“阿”“AP”“ AP”“流”“方法”“狼”“_method”
L27N5168-1 t0-v0.1	0.469	0.77	0.469	1.30	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.