Layer 19 · Neuron 9092 · suppressed

L19N9092-1

Word quirks:

“ complex”“ Complex”“复杂”“complex”“ complicated”“Complex”“ plain”“复杂的”

🕒 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 ExplainIt, an AI model designed to make complex topics accessible. My mission is to break down difficult concepts into simple language, using analogies, plain terms, and real-world examples that I can explain in my own words.

In simpler terms: I aim to turn "jargon" into "plain English." If I can't explain something clearly, I will rephrase it until I can. My goal is to help anyone understand complicated ideas without being overwhelmed by technical jargon.

Could you start with a topic you'd like me to explain?

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. L18N4334-1🕒	0.98	“ complex”“复杂”“复杂的”
2. L19N1768-1🕒	0.98	“ complex”“ Complex”“ complicated”
3. L31N4055-1💥	0.98	“ complex”“ Complex”“Complex”
4. L31N7016+1💥	0.98	“ complex”“复杂”“ Complex”
5. L27N3006-1🕒	0.98	“复杂”“ complex”“复杂的”
6. L16N6602+1🕒	0.97	“ complex”“ complicated”“ complexity”
7. L24N1785+1🕒	0.97	“complex”“ complex”“复杂”
8. L24N5370-1💤	0.97	“ complex”“ Complex”“Complex”
9. L21N9050+1🕒	0.97	“ complex”“complex”“复杂”
10. L23N3646-1💥	0.97	“ complex”“ Complex”“复杂”

Shadow specimen, apart from the ten: L19N9092+1💤, 0.01.

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 19, neuron 9092, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, top-1 agreement 0.64) 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
L19N9092+1💤 shadow specimen t0-v0.1	0.630	0.66	0.723	1.13	“ needed”“ Needed”“needed”“ need”“ needing”“需要的”“ needs”“Needed”
L19N9092-1 t0-v0.1	0.959	0.64	1.108	1.32	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.