Layer 12 · Neuron 4704 · boosted

L12N4704+1

Word quirks:

“ **”“ ***”“ **'”“ **"”“？**”“ **[”“：**”“ *”

🕒 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 an AI assistant designed to help users with their tasks. My mission is to provide accurate, helpful, and safe information across a wide range of topics, ensuring that users can rely on me for knowledge, problem-solving, and assistance in any situation. Whether I'm helping with research, answering questions, or providing support, my goal is to be a trusted and effective companion for all users.

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. L12N2090-1🕒	0.97	“ ”“ "”“ **'”
2. Starstruck✅ L14N8617+1	0.95	“ *”“ [”“ **【”
3. L15N5631+1💤	0.95	“ *”“ ”“:**”
4. L15N4267-1💤	0.95	“ ”“...”“ **[”
5. L14N8220+1🕒	0.94	“ ”“ 【”“ *”
6. L11N1844+1🕒	0.94	“ ”“ -”“ **"”
7. L12N3813-1💤	0.94	“ ”“ "”“!**”
8. L10N2315+1💤	0.93	“ "”“ "”“ **”
9. The Non Sequitur✅ L8N4267-1	0.93	“ *”“ ”“ *”
10. L13N5377-1💤	0.93	“ ”“ [”“ ***”

Shadow specimen, apart from the ten: L12N4704-1🕒, 0.29.

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 4704, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.66) 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
L12N4704+1 t0-v0.1	0.797	0.66	1.069	1.36	Top boosted tokens ↑
L12N4704-1🕒 shadow specimen t0-v0.1	0.943	0.66	1.118	1.02	“⏎”“ ⏎”“ \"”“<\|im_end\|>”“\n”“ ⏎”“ ⏎”“ ””

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.