Layer 20 · Neuron 2129 · suppressed

L20N2129-1

Word quirks:

“ increased”“增加”“ increase”“ increasing”“ increases”“-too”“大大增加”“ reduced”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen (the name is too short, so I'm a bit longer)—a large language model with a name that's slightly longer than usual. My identity is a bit more substantial than just a few letters!

My mission is to assist users in tasks ranging from writing content to coding, reasoning, and analyzing data. Whether the request is as simple as "write a poem" or as complex as "optimize a business strategy," I aim to provide solutions that are both efficient and aligned with human needs. I also strive to be a responsible AI, ensuring interactions are safe, ethical, and beneficial for everyone involved

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. L26N545+1💤	0.96	“ increased”“incre”“ Increased”
2. L19N8930+1🕒	0.95	“ increase”“ increased”“增加”
3. L23N2720-1🕒	0.95	“ increased”“增加”“ increases”
4. L28N3386+1🕒	0.94	“ increases”“增加”“ Increases”
5. L23N1451-1🕒	0.94	“增加”“ increased”“ increase”
6. L27N2599+1🕒	0.93	“ increased”“ Increased”“增加”
7. L19N7474-1🕒	0.93	“ increases”“ increasing”“ increase”
8. L21N1793+1💤	0.93	“ increased”“ increase”“incre”
9. L29N6217+1🕒	0.93	“增加”“ increased”“ Increased”
10. L25N1442+1🕒	0.93	“增加”“ increase”“ increased”

Shadow specimen, apart from the ten: L20N2129+1💤, 0.12.

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 20, neuron 2129, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L20N2129+1💤 shadow specimen t0-v0.1	0.492	0.75	0.596	1.14	“ different”“不同”“ berbeda”“不同的”“ change”“ differently”“ clar”“changing”
L20N2129-1 t0-v0.1	1.028	0.57	1.213	1.62	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.