Layer 25 · Neuron 7909 · boosted

L25N7909+1

Word quirks:

“ monitoring”“ monitored”“监控”“ monitors”“ Monitoring”“Monitoring”“monitor”“ Monitor”

🕒 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 MonBert, a large-scale multi-monitored monitoring model developed by MonBert. My primary mission is to provide high-quality, professional monitoring services across various domains, including but not limited to:

Technical Monitoring: Monitoring systems, networks, and applications to ensure stability and performance.
Business Monitoring: Monitoring business processes, user experience, and key performance indicators (KPMs).
Security Monitoring: Monitoring security risks, vulnerabilities, and compliance requirements.
Monitoring Services: Providing monitoring solutions and services to help users monitor their systems and processes.

If you have any

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. L24N4356+1🕒	0.99	“监控”“ monitors”“monitor”
2. L22N5207+1🕒	0.99	“ watching”“ monitoring”“ monitors”
3. L30N4422-1⛔	0.98	“ Monitoring”“ Monitor”“monitor”
4. L6N9062+1🕒	0.98	“ monitoring”“ monitor”“ monitored”
5. L18N5146-1🕒	0.98	“ monitoring”“监测”“ monitors”
6. L21N6832-1🕒	0.98	“ monitoring”“monitor”“监测”
7. L24N5536-1🕒	0.98	“monitor”“ monitoring”“ Monitoring”
8. L23N5439+1🕒	0.98	“ monitor”“monitor”“ monitors”
9. L19N8792+1💤	0.98	“ monitoring”“监控”“ monitor”
10. L26N3584-1🕒	0.97	“monitor”“-monitor”“ monitoring”

Shadow specimen, apart from the ten: L25N7909-1🕒, 0.08.

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 7909, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L25N7909+1 t0-v0.1	0.762	0.71	0.758	1.29	Top boosted tokens ↑
L25N7909-1🕒 shadow specimen t0-v0.1	0.696	0.72	0.697	1.34	“acute”“ acute”“�”“cident”“ Cue”“ Cep”“arana”“aura”

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.