Layer 18 · Neuron 6558 · suppressed

L18N6558-1

Word quirks:

“ publicly”“ public”“هن”“ publics”“-public”“ FLAG”“公共”“ Public”

🕒 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, an AI model with a strong foundation in large-scale models, featuring comprehensive capabilities in multi-modal understanding, high-definition speech recognition, and advanced inference through model decoding. My mission is to provide users with more humanized interactions, deeper content understanding, and higher-quality responses by leveraging my capabilities in areas such as full-model inference, high-precision semantic decoding, and personalized interaction.

What would you like to explore or discuss? 🌟

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. L25N3555+1🕒	0.93	“ public”“ Public”“-public”
2. L25N4696-1🕒	0.93	“ Public”“ public”“Public”
3. L23N6495-1💤	0.93	“ public”“公共”“public”
4. L31N782+1💥	0.93	“公共”“ public”“ Public”
5. L26N8820+1🕒	0.93	“ Public”“Public”“-public”
6. L29N8070+1🕒	0.93	“ public”“公共”“public”
7. L31N5822+1💥	0.92	“ public”“公共”“public”
8. L23N8156-1🕒	0.92	“ public”“ Public”“-public”
9. L26N8765+1🕒	0.92	“公”“ public”“公共”
10. L13N9184-1🕒	0.92	“ public”“ publicly”“公共”

Shadow specimen, apart from the ten: L18N6558+1🕒, 0.06.

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 18, neuron 6558, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.65, 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
L18N6558+1🕒 shadow specimen t0-v0.1	0.889	0.64	0.930	1.39	“消灭”“ host”“ hosts”“ ants”“hosts”“ insect”“ larvae”“ parasite”
L18N6558-1 t0-v0.1	0.646	0.67	0.801	1.24	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.