Layer 20 · Neuron 4980 · suppressed

L20N4980-1

Word quirks:

“ instant”“ instantaneous”“ index”“ instantiated”“ instantiation”“-$”“ indices”“针对性”

🕒 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 Qwen, a large model trained by the Tianzi Model (Tianzi Model). My mission is to provide high-quality, accurate, and helpful responses across various domains, including knowledge, reasoning, code generation, and personalized assistance. I aim to be a reliable resource for users, ensuring that my interactions are clear, precise, and tailored to their needs. Whether you need to solve complex problems, create content, or simply ask a question, I strive to deliver the best possible support. How can I assist you?

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. L2N5721+1💥	0.73	“ Instance”“ instance”“<\|im_end\|>”
2. L9N3847-1🕒	0.71	“⏎”“ activation”“ instantiated”
3. L25N5683+1🕒	0.71	“ instant”“ Instant”“instant”
4. L25N2089+1🕒	0.70	“ instant”“ Instant”“Instant”
5. L21N1472-1💤	0.68	“实例”“ instance”“ instances”
6. L24N4911-1🕒	0.68	“ instant”“ instantaneous”“instant”
7. L27N7795-1💤	0.68	“ instant”“ ev”“instant”
8. L24N6399-1🕒	0.68	“ instant”“ Instant”“ instantaneous”
9. L24N3471-1🕒	0.67	“ instantaneous”“ Instant”“Instant”
10. L29N3839-1💥	0.67	“ instant”“instant”“ instantaneous”

Shadow specimen, apart from the ten: L20N4980+1🕒, 0.10.

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 4980, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.69) 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
L20N4980+1🕒 shadow specimen t0-v0.1	1.179	0.63	1.332	1.40	“ Indonesia”“福建”“ Japan”“浙江”“江苏”“ China”“四川”“中国”
L20N4980-1 t0-v0.1	0.696	0.69	0.832	1.20	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.