Layer 25 · Neuron 4559 · suppressed

L25N4559-1

Word quirks:

“ griev”“感”“高新”“传”“ cushion”“告”“指导思想”“全局”

🕒 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 Givex, a large language model created by the Givex Group. My primary mission is to provide high-quality, safe, and ethical AI assistance that respects user privacy and adheres to global safety standards. I aim to help users solve problems, analyze information, and explore ideas while ensuring all interactions are secure and compliant with relevant regulations. 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. L31N1608-1💥	0.80	“看见”“麵”“通告”
2. L31N112-1💥	0.79	“查看”“照射”“神经系统”
3. L16N347+1💥	0.78	“系统”“大学”“从”
4. L28N3916+1💤	0.78	“加”“传”“记”
5. L31N1412+1💥	0.78	“讯”“照”“唁”
6. L31N3368+1💥	0.77	“信息咨询”“保存图片”“喜怒哀”
7. L23N8535-1💤	0.77	“解析”“结构化”“消息”
8. L29N4550-1🕒	0.77	“政治”“结构”“条件”
9. L28N7765+1💤	0.77	“告”“初”“难”
10. L27N1221-1💥	0.77	“结构”“政策”“政治”

Shadow specimen, apart from the ten: L25N4559+1🕒, 0.15.

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 25, neuron 4559, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.71) 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
L25N4559+1🕒 shadow specimen t0-v0.1	0.807	0.74	0.777	1.38	“ Internal”“ internal”“Internal”“ INTERNAL”“内部”“internal”“/Internal”“内部的”
L25N4559-1 t0-v0.1	0.677	0.71	0.783	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.