Layer 7 · Neuron 3983 · boosted

L7N3983+1

Word quirks:

“ travel”“ route”“ roadside”“ Route”“ road”“ highway”“ roads”“ routes”

🕒 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 an AI assistant designed to help you with navigation, rest stops, or quick errands. My "mission" is to provide a break from the road, offering services like food, coffee, restroom access, and sometimes local shopping or sightseeing. Whether you're traveling for work, vacation, or just need a change of pace, I'm here to assist!

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. L6N4258+1🕒	0.98	“ Route”“ route”“ roads”
2. L7N6249+1🕒	0.97	“ roads”“ route”“ routes”
3. L4N5755-1🕒	0.97	“ roads”“ road”“ route”
4. L7N6499+1🕒	0.97	“ roads”“ routes”“ route”
5. L7N6185+1🕒	0.97	“ roads”“ road”“ route”
6. L8N8691+1🕒	0.96	“ road”“ roads”“ route”
7. L2N6153+1💥	0.94	“ road”“ roads”“<\|im_end\|>”
8. L22N8565+1💥	0.93	“ route”“ path”“ Route”
9. L5N7901+1🕒	0.93	“ roads”“ highway”“ Highway”
10. L5N5282-1💥	0.93	“ highway”“ freeway”“ route”

Shadow specimen, apart from the ten: L7N3983-1🕒, 0.08.

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 7, neuron 3983, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.40, top-1 agreement 0.49) 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
L7N3983+1 t0-v0.1	1.401	0.49	1.672	1.75	Top boosted tokens ↑
L7N3983-1🕒 shadow specimen t0-v0.1	1.564	0.55	1.853	1.43	“ DNS”“ domains”“ https”“ domain”“ hosts”“ http”“ hostname”“ www”

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.