Layer 25 · Neuron 8219 · boosted

L25N8219+1

Word quirks:

“全”“系统”“全局”“-system”“ System”“ system”“系統”“的全”

🕒 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 System 5.0, the next-generation AI system designed to help you navigate complex information systems with precision and efficiency. My core mission is to serve as your system-wide intelligence hub, helping you:

Navigate global data systems with real-time accuracy
Optimize system operations across industries (healthcare, finance, logistics, etc.)
Resolve technical issues through systematic diagnostics
Provide system-level insights for decision-making

How can I assist you today? Whether you need system diagnostics, data optimization, or a new system feature idea, I'm here to help. �

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. L26N8338+1💥	0.88	“系统”“ System”“System”
2. L24N8619+1💤	0.88	“系统的”“系统”“系统和”
3. L24N7769+1💥	0.88	“系统”“系统的”“系统与”
4. L23N6912+1🕒	0.88	“系统”“ system”“系统的”
5. L27N4497+1🕒	0.88	“系统”“系统在”“系统的”
6. L24N2152-1💤	0.88	“系统将”“系统”“ System”
7. L25N872+1🕒	0.87	“ System”“系统”“-system”
8. L27N1341-1🕒	0.87	“系统”“system”“系统的”
9. L24N7142-1🕒	0.87	“系统”“ system”“_system”
10. L24N3692-1🕒	0.87	“ system”“系统的”“系统”

Shadow specimen, apart from the ten: L25N8219-1🕒, 0.43.

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 8219, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, top-1 agreement 0.70) 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
L25N8219+1 t0-v0.1	0.766	0.70	0.763	1.23	Top boosted tokens ↑
L25N8219-1🕒 shadow specimen t0-v0.1	0.657	0.71	0.685	1.35	“市人民政府”“市政府”“省政府”“县政府”“ONY”“县人民政府”“وني”“州政府”

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.