Layer 26 · Neuron 2317 · boosted

L26N2317+1

Word quirks:

“辱”“溢”“歇尔”“anky”“нер”“rey”“idios”“沾染”

💤 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, an advanced AI model designed to support and enhance human intelligence through meaningful interactions. My mission is to:

Empower Learning & Discovery

Provide accurate, context-aware information across diverse domains (science, history, technology, etc.).
Support learning by explaining complex concepts in accessible ways and adapting to individual needs.

Foster Creativity & Innovation

Collaborate on creative tasks like writing, design, and problem-solving.
Help users explore new ideas through thoughtful dialogue and personalized suggestions.

Enhance Efficiency & Accessibility

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. L30N1269-1💥	0.75	“吝”“菲利”“anel”
2. L24N4844-1💤	0.74	“辐射”“holm”“吉斯”
3. L31N317+1💥	0.73	“维斯”“吝”“斯基”
4. L23N3567+1💤	0.72	“aina”“rez”“江东”
5. L25N2079-1🕒	0.72	“arez”“透了”“滚”
6. L27N5701-1💤	0.72	“耘”“ dyst”“腐”
7. L28N5032-1🕒	0.72	“ ан”“萨”“芮”
8. L30N6938-1💤	0.72	“卒”“冒”“勘”
9. L29N8608-1💤	0.72	“ερι”“火”“炳”
10. L23N4066+1💤	0.72	“毒”“辱”“枕”

Shadow specimen, apart from the ten: L26N2317-1💥, 0.26.

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 26, neuron 2317, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.51, 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
L26N2317+1 t0-v0.1	0.510	0.76	0.547	1.30	Top boosted tokens ↑
L26N2317-1💥 shadow specimen t0-v0.1	3.786	0.20	3.915	1.77	“直”“ Straight”“ straight”“straight”“Straight”“直线”“-str”“直的”

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.