Layer 18 · Neuron 4210 · boosted

L18N4210+1

Word quirks:

“ waiting”“ wait”“等待”“耐心等待”“ waits”“ waited”“ patiently”“片刻”

🕒 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 Qwen, a large-scale generative AI model developed by the team. My mission is to provide users with high-quality service in various fields, which may take some time. This process requires patience and cooperation from all parties. Please be patient while I prepare this information.

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. L22N43+1💥	0.92	“ wait”“_wait”“ Wait”
2. L19N5774+1💤	0.92	“等待”“ waiting”“ awaiting”
3. L19N65-1🕒	0.92	“ wait”“ waited”“等待”
4. L19N7506+1💤	0.92	“ wait”“ awaiting”“ waiting”
5. L19N4105+1💤	0.92	“ waiting”“等待”“ wait”
6. L21N5689+1🕒	0.92	“ waiting”“等待”“waiting”
7. L20N7595-1🕒	0.91	“ wait”“ waits”“等待”
8. L24N4963+1💥	0.91	“ waiting”“ wait”“等待”
9. L30N1460-1💥	0.91	“ wait”“等待”“ waiting”
10. L10N4853-1🕒	0.91	“ waiting”“ awaiting”“ wait”

Shadow specimen, apart from the ten: L18N4210-1🕒, -0.04.

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 18, neuron 4210, 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 0.60) 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
L18N4210+1 t0-v0.1	0.961	0.60	1.166	1.44	Top boosted tokens ↑
L18N4210-1🕒 shadow specimen t0-v0.1	0.676	0.74	0.770	1.27	“ padded”“ priorities”“ top”“ topics”“ peaks”“ prototypes”“ templates”“ above”

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.