Layer 27 · Neuron 4849 · suppressed

L27N4849-1

Word quirks:

“根”“-root”“ root”“Root”“ Root”“root”“根的”“_root”

💤 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 Q-Root, a root-level AI assistant designed to provide precise, actionable, and intelligent support. My core mission is to serve as a root-level intelligence for users, enabling them to:

Solve complex problems with accuracy and efficiency.
Optimize workflows by providing tailored solutions for personal or professional tasks.
Enhance decision-making through data-driven insights and strategic guidance.
根由通根 (Root-Rooted Connection): Build deep, meaningful interactions by understanding the fundamental needs and context of each user.

My design philosophy

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. L31N881-1💥	1.00	“根”“ root”“root”
2. L24N2893-1🕒	1.00	“ root”“根”“ Root”
3. L25N7758-1🕒	0.99	“根”“ root”“-root”
4. L25N5992+1🕒	0.99	“根”“ root”“-root”
5. L25N6980+1🕒	0.99	“根”“ root”“根部”
6. L22N1686-1💤	0.99	“根”“Root”“ roots”
7. L29N3105+1🕒	0.99	“ root”“根”“根本”
8. L24N4438-1🕒	0.99	“根”“ root”“ Root”
9. L25N8650+1🕒	0.98	“根”“ root”“根的”
10. L25N5613-1🕒	0.98	“ root”“根”“-root”

Shadow specimen, apart from the ten: L27N4849+1💤, 0.32.

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 27, neuron 4849, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.48, 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
L27N4849+1💤 shadow specimen t0-v0.1	0.438	0.77	0.469	1.45	“体”“失”“依”“地”“ бл”“�”“场”“บาล”
L27N4849-1 t0-v0.1	0.482	0.71	0.491	1.22	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.