Layer 19 · Neuron 4757 · boosted

L19N4757+1

Word quirks:

“ completed”“ finished”“ concluded”“ released”“ finishing”“ Finished”“ Completed”“ completion”

🕒 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 10 released by Alibaba Cloud. My mission is to empower users with advanced capabilities in natural language processing, multi-modal analysis, and complex problem-solving. Whether you need help with creative writing, logical reasoning, data analysis, or long-term interactions, I aim to provide efficient, accurate, and personalized assistance. Let's collaborate on a wide range of tasks! 🚀

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. The Finalizer✅ L23N3529+1	0.94	“ finished”“ completed”“ completion”
2. L21N5631-1🕒	0.92	“ finished”“ completed”“完成的”
3. Done and Done✅ L23N3529+1,L25N2453-1	0.92	“已完成”“ finished”“ completed”
4. L10N9211+1🕒	0.92	“ completed”“ completion”“ final”
5. L12N8196+1🕒	0.91	“ completion”“ completing”“ finishing”
6. L19N5309+1🕒	0.91	“ completion”“ completed”“ submission”
7. L19N7921+1🕒	0.91	“ completion”“ complet”“ completing”
8. L16N8196-1🕒	0.91	“ completion”“ completes”“ completing”
9. L13N4796+1🕒	0.90	“ completion”“ final”“ completed”
10. L20N6808-1🕒	0.90	“ finishing”“ finish”“ completion”

Shadow specimen, apart from the ten: L19N4757-1💤, 0.07.

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 4757, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.66, top-1 agreement 0.73) 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
L19N4757+1 t0-v0.1	0.665	0.73	0.792	1.40	Top boosted tokens ↑
L19N4757-1💤 shadow specimen t0-v0.1	0.498	0.74	0.596	1.17	“ current”“ Current”“ core”“ according”“current”“.current”“(current”“Current”

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.