Layer 28 · Neuron 2896 · boosted

L28N2896+1

Word quirks:

“uess”“abler”“öl”“ogia”“OF”“olia”“询价”“ож”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Q3, an AI assistant developed by T (T) to provide you with the most accurate and helpful information possible. My mission is to assist you in solving problems, answering questions, and creating content across a wide range of topics—whether that's writing code, analyzing data, generating creative ideas, or simply having an engaging conversation.

How can I help 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. L29N8573-1💤	0.75	“odi”“oly”“ol”
2. L30N7553+1💤	0.74	“op”“olie”“aten”
3. L28N5082-1🕒	0.73	“lig”“ol”“ оли”
4. L29N7751-1🕒	0.72	“olina”“olini”“olly”
5. L31N7463+1💥	0.72	“oz”“ox”“ostal”
6. L29N5674+1💤	0.72	“olio”“olut”“jew”
7. L31N8362-1💥	0.71	“oker”“ocl”“ož”
8. L29N8953-1🕒	0.71	“ol”“olo”“ole”
9. L31N3172+1💥	0.70	“o”“oS”“oz”
10. L27N1314-1💤	0.70	“银行存款”“oun”“okit”

Shadow specimen, apart from the ten: L28N2896-1🕒, 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 28, neuron 2896, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.43, 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
L28N2896+1 t0-v0.1	0.434	0.76	0.515	1.48	Top boosted tokens ↑
L28N2896-1🕒 shadow specimen t0-v0.1	0.457	0.79	0.493	1.30	“ total”“ Total”“总”“Total”“-total”“total”“(total”“_total”

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.