Layer 11 · Neuron 4212 · boosted

L11N4212+1

Word quirks:

“ **"”“ "**”“ your”“ **“”“⏎”“ *"”“ "”“ you”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, the large multimodal model series from Alibaba. My mission is to provide high-quality services for users like you. I will continue to support and assist you in various tasks, ensuring that your interactions with me are smooth and efficient. How can I help you?

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. L12N2722-1💤	0.95	“ “”“ you”“ "”
2. L15N882-1🕒	0.94	“ "”“ "”“ *"”
3. L8N5469+1💤	0.93	“ "”“ “”“ "”
4. L12N8873-1🕒	0.93	“ "”“ your”“ "”
5. L7N1530-1💤	0.92	“ “”“ "”“ "**”
6. L16N3246+1💤	0.91	“—"”“ "”“ “”
7. L8N7605-1🕒	0.91	“ "”“ “”“ your”
8. L10N8543+1🕒	0.91	“ `"”“ your”“ *"”
9. L15N5522-1💤	0.91	“⏎”“ **“”“ your”
10. L15N2052-1💤	0.90	“ “”“ "”“ **„”

Shadow specimen, apart from the ten: L11N4212-1🕒, 0.32.

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 11, neuron 4212, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.69, top-1 agreement 0.69) 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
L11N4212+1 t0-v0.1	0.691	0.69	0.859	1.18	Top boosted tokens ↑
L11N4212-1🕒 shadow specimen t0-v0.1	2.092	0.43	2.549	2.58	“ you”“?”“ these”“？”“ what”“ ?”“ your”“?\”

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.