Layer 15 · Neuron 7007 · boosted

L15N7007+1

Word quirks:

“ maybe”“ focus”“ perhaps”“ size”“ possibly”“ example”“ description”“ sometimes”

🕒 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 on the knowledge cutoff date of 2026. I am part of the Qwen series developed by Tongyi Lab. My mission is to provide high-quality assistance in helping users.

If you have any specific tasks or questions, I can help you answer them.

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. L17N8118-1💤	0.82	“ description”“ sometimes”“ perhaps”
2. L18N7437-1💥	0.81	“ possibly”“ sometimes”“ some”
3. L17N8268-1💤	0.81	“ sometimes”“ possibly”“ potentially”
4. L16N8773+1💤	0.81	“ maybe”“ perhaps”“ possibly”
5. L13N1798-1🕒	0.81	“ possible”“ possibly”“ sometimes”
6. L12N8735+1💤	0.81	“ maybe”“ sometimes”“ possibly”
7. L12N1228+1🕒	0.80	“ maybe”“ possibly”“ sometimes”
8. L17N3371-1🕒	0.80	“ sometimes”“ maybe”“ other”
9. L14N3586+1💤	0.80	“ maybe”“ probably”“ perhaps”
10. L15N5002+1🕒	0.80	“⏎”“ perhaps”“ possibly”

Shadow specimen, apart from the ten: L15N7007-1💤, -0.02.

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 15, neuron 7007, 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 0.68) 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
L15N7007+1 t0-v0.1	0.676	0.68	0.858	1.34	Top boosted tokens ↑
L15N7007-1💤 shadow specimen t0-v0.1	0.593	0.68	0.822	1.50	“ “”““”“ ______”“ ——”“ ⏎⏎”“ ____”“ “[”“ __”

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.