Layer 12 · Neuron 4545 · boosted

L12N4545+1

Word quirks:

“<|im_end|>”“ same”“ seperate”“ separate”“ similarly”“ ***”“ Same”“ similar”

💤 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 latest large language model in the Qwen series developed by Tongyi Lab. My mission is to provide high-quality, intelligent, and helpful services that support human development and contribute to societal progress. I aim to assist users in solving problems, enhancing capabilities, and promoting positive outcomes through advanced AI technologies.

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. L16N4062+1🕒	0.88	“ Same”“ same”“ similar”
2. L13N7989-1🕒	0.87	“ same”“ similar”“<\|im_end\|>”
3. L12N6995-1🕒	0.85	“ similar”“ similarly”“ separate”
4. L13N5308+1💤	0.85	“ similar”“ separate”“ separately”
5. L11N534+1💤	0.85	“ similarly”“ separate”“ similar”
6. L16N4661+1💤	0.85	“ similar”“ same”“ equally”
7. L15N2814-1🕒	0.84	“ same”“ uniformly”“ actually”
8. L10N8289-1🕒	0.84	“ similar”“ similarly”“ same”
9. L12N5951+1🕒	0.84	“ same”“ respective”“ Same”
10. L14N6656+1💤	0.84	“ same”“ different”“ Same”

Shadow specimen, apart from the ten: L12N4545-1💤, 0.09.

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 12, neuron 4545, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L12N4545+1 t0-v0.1	0.756	0.67	0.911	1.15	Top boosted tokens ↑
L12N4545-1💤 shadow specimen t0-v0.1	0.661	0.67	0.811	1.16	“ actually”“ actual”“ example”“ first”“ defining”“ defined”“ generally”“ overall”

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.