Layer 22 · Neuron 1686 · suppressed

L22N1686-1

Word quirks:

“根”“Root”“ roots”“-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 Qianse, an AI assistant developed by Qianhai Group. My primary goal is to provide high-quality information, answer questions, and support decision-making processes across various industries. I aim to deliver accurate, reliable, and timely insights while adhering to the principles of safety and compliance.

As a large language model, I can assist with tasks such as:

Explaining complex topics in clear, concise ways.
Generating content for creative or professional use.
Analyzing data or providing industry-specific insights.
Supporting research, documentation, or workflow automation.

My mission is to empower users

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. L25N5992+1🕒	0.99	“根”“ root”“-root”
2. L25N7758-1🕒	0.99	“根”“ root”“-root”
3. L25N5613-1🕒	0.99	“ root”“根”“-root”
4. L25N6980+1🕒	0.99	“根”“ root”“根部”
5. L24N2893-1🕒	0.99	“ root”“根”“ Root”
6. L27N4849-1💤	0.99	“根”“-root”“ root”
7. L25N4328-1💥	0.99	“根”“ root”“ roots”
8. L26N1709+1🕒	0.98	“ root”“根”“ roots”
9. L25N5794+1🕒	0.98	“根”“ root”“ roots”
10. L31N881-1💥	0.98	“根”“ root”“root”

Shadow specimen, apart from the ten: L22N1686+1💤, 0.34.

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 22, neuron 1686, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, 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
L22N1686+1💤 shadow specimen t0-v0.1	0.950	0.66	1.160	1.55	“马鞍”“im”“ ped”“鞍”“-app”“ Format”“上身”“ Dot”
L22N1686-1 t0-v0.1	0.743	0.70	0.792	1.28	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.