Layer 16 · Neuron 7080 · suppressed

L16N7080-1

Word quirks:

“ similar”“ same”“ needed”“ exactly”“ describing”“ your”“ similarly”“ like”

🕒 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 Qwen3.5, a large language model trained by Alibaba Cloud. My mission is to assist users in generating text, code, or reasoning tasks with high accuracy and efficiency. I strive to provide clear, concise, and helpful responses across various domains, whether I'm solving math problems, writing stories, or analyzing data. If you need help, feel free to ask! 😊

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. L15N2891+1🕒	0.87	“ similar”“ same”“ similarly”
2. L14N5198+1💤	0.87	“ similar”“ similarly”“ same”
3. L18N2578+1💤	0.86	“ similar”“ like”“ your”
4. L12N9027-1💤	0.86	“ similarly”“ similar”“ same”
5. L14N6326-1💤	0.86	“ similarly”“ similar”“ this”
6. L12N6337+1🕒	0.86	“ similarly”“ similar”“ likewise”
7. L15N5627-1🕒	0.85	“ similarly”“ similar”“ same”
8. L15N7994+1🕒	0.85	“ similar”“ similarly”“ same”
9. L16N2339-1💤	0.85	“ just”“ similar”“ akin”
10. L13N6586+1🕒	0.85	“ similar”“ similarly”“ same”

Shadow specimen, apart from the ten: L16N7080+1💤, -0.12.

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 16, neuron 7080, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.67) 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
L16N7080+1💤 shadow specimen t0-v0.1	0.523	0.72	0.718	1.33	“ “”“ ‘”“ ‘’”“ said”“ “[”“ \"”“ famed”“ here”
L16N7080-1 t0-v0.1	0.678	0.67	0.833	1.07	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.