Layer 24 · Neuron 4635 · suppressed

L24N4635-1

Word quirks:

“纳”“计”“iau”“霓”“ENE”“/U”“吉”“ neger”

💤 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 G60, a specialized AI assistant developed to support the G60 Innovation Action Zone in the Chinese National Private Sector. My primary mission is to act as a strategic partner for the private sector by:

Deepening Digital Transformation: Helping enterprises integrate digital tools to improve efficiency, innovation, and competitiveness.
Enhancing Public-Private Collaboration: Faciating seamless interactions between government resources and private sector needs to solve complex challenges.
Driving Regional Innovation: Supporting the G60 Action Zone's goal of becoming a global hub for private sector innovation through targeted insights and actionable strategies

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. L28N9105+1💤	0.81	“纳”“0”“ssel”
2. L27N3410-1💤	0.79	“纳”“米”“厦”
3. L21N9+1💥	0.78	“ Seymour”“ิน”“欧”
4. L26N1237-1🕒	0.78	“chnik”“切”“纳”
5. L27N8429+1💤	0.78	“凌”“納”“取”
6. L29N9048-1🕒	0.78	“达”“仪”“诺”
7. L28N8142-1🕒	0.78	“尼”“ Ney”“ н”
8. L28N3702-1💤	0.77	“纳”“硕”“惠”
9. L30N2003+1💤	0.77	“阿”“计”“eno”
10. L24N3536-1💤	0.77	“ment”“.bi”“nj”

Shadow specimen, apart from the ten: L24N4635+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 24, neuron 4635, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.72) 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
L24N4635+1🕒 shadow specimen t0-v0.1	2.211	0.50	2.297	1.13	“ bottom”“ Bottom”“bottom”“ top”“Bottom”“-bottom”“-top”“ bottoms”
L24N4635-1 t0-v0.1	0.750	0.72	0.847	1.37	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.