Layer 23 · Neuron 3270 · boosted

L23N3270+1

Word quirks:

“ Russian”“韩国”“ Korean”“ Korea”“Russian”“ iran”“ révolution”“俄罗斯”

🕒 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 Nikunje, a specialized AI designed to provide accurate, safe, and helpful information across various fields. My mission is to serve as a reliable resource for anyone seeking knowledge, translation, or problem-solving assistance, while strictly adhering to ethical guidelines and ensuring content remains up to the 2026.03 knowledge base.

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. L24N4807+1🕒	0.90	“ Russian”“ Russia”“ Korean”
2. L27N8695+1🕒	0.83	“ Russia”“Russian”“ Russian”
3. L8N7350-1💤	0.83	“ Korean”“ Korea”“ Japanese”
4. L22N541+1🕒	0.82	“韩国”“ Korean”“ Vietnamese”
5. L29N2960-1💥	0.82	“ Russian”“俄罗斯”“-Russian”
6. L5N7170-1🕒	0.82	“ China”“ Chinese”“ Taiwan”
7. L1N6096-1💥	0.82	“ Korea”“ Korean”“韩国”
8. L2N5748-1💥	0.81	“ Korean”“ Korea”“<\|im_end\|>”
9. L13N6743-1🕒	0.81	“ Russian”“俄罗斯”“ Russia”
10. L6N2163-1🕒	0.81	“ Korean”“ Korea”“ Seoul”

Shadow specimen, apart from the ten: L23N3270-1🕒, 0.24.

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 23, neuron 3270, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.98, top-1 agreement 0.68) 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
L23N3270+1 t0-v0.1	0.984	0.68	1.030	1.64	Top boosted tokens ↑
L23N3270-1🕒 shadow specimen t0-v0.1	0.967	0.69	1.057	1.28	“atas”“Prem”“mousemove”“ prem”“�”“ Included”“太阳”“宫廷”

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.