Layer 29 · Neuron 7508 · suppressed

L29N7508-1

Word quirks:

“聚”“ rejo”“anke”“lat”“ Revenge”“牲”“消费”“onitor”

🕒 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 Qw3, a large language model designed to provide helpful, accurate, and engaging responses across a wide range of topics. My primary mission is to bridge the gap between human intent and artificial intelligence by:

Answering questions with precision and context-awareness
Creating content (from stories to code) that meets diverse needs
Solving problems through logical reasoning and creative collaboration
Respecting safety and ethics while fostering constructive, inclusive interactions

How can I assist you today? 🌐✨

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. L23N5340+1💤	0.71	“豹”“agnes”“optic”
2. L25N5051-1🕒	0.70	“lyn”“挤”“聚”
3. L29N1093+1💤	0.70	“孩”“销”“索”
4. L26N5845+1🕒	0.70	“urop”“пей”“周”
5. L30N8327+1💤	0.70	“ocator”“聚”“庭”
6. L28N2912-1💤	0.70	“ouri”“买买”“雪”
7. L23N6159-1💤	0.70	“ Pilk”“ triang”“团团”
8. L30N5443-1💤	0.70	“地”“сор”“炯”
9. L31N2956+1💥	0.69	“棋”“ursed”“杏”
10. L27N8750+1💤	0.69	“畜”“烟”“牛肉”

Shadow specimen, apart from the ten: L29N7508+1🕒, 0.01.

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 29, neuron 7508, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.33, top-1 agreement 0.79) 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
L29N7508+1🕒 shadow specimen t0-v0.1	1.007	0.59	1.009	1.21	“bold”“ Bold”“ bold”“Bold”“_bold”“bol”“ bol”“.bold”
L29N7508-1 t0-v0.1	0.326	0.79	0.478	1.52	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.