Layer 20 · Neuron 4175 · boosted

L20N4175+1

Word quirks:

“ lesser”“ smaller”“ weaker”“ Lesser”“MASTER”“ shorter”“ less”“ genius”

🕒 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, developed by Qianyuan (the AI giant) as a large language model. My mission is to assist users by providing accurate information, enabling creative problem-solving, and enhancing efficiency through intelligent interaction. Whether I need to help with data analysis, content creation, logical reasoning, or simply having a conversation, my goal is to be a capable and respectful partner in making life easier for everyone. How can I help you right now? 😊

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. L19N7773+1🕒	0.89	“ shorter”“ less”“ smaller”
2. L18N6723+1🕒	0.88	“ fewer”“ smaller”“ weaker”
3. L19N5804-1💤	0.86	“ lesser”“ less”“ menos”
4. L21N5822+1🕒	0.86	“ whichever”“ poorer”“ lesser”
5. L22N3230+1🕒	0.86	“ lesser”“ poorer”“ larger”
6. L15N5252+1🕒	0.85	“ less”“ fewer”“ quieter”
7. L2N4279-1💥	0.85	“ less”“ Less”“ than”
8. L15N4148-1🕒	0.85	“ smaller”“ simpler”“ without”
9. L16N7721-1🕒	0.84	“ less”“ smaller”“ simpler”
10. L17N4117-1🕒	0.84	“ less”“ lesser”“ unchecked”

Shadow specimen, apart from the ten: L20N4175-1🕒, 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 4175, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, 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
L20N4175+1 t0-v0.1	0.714	0.69	0.840	1.25	Top boosted tokens ↑
L20N4175-1🕒 shadow specimen t0-v0.1	1.431	0.61	1.531	1.37	“ year”“ millennium”“ Year”“ mill”“ new”“年新”“ century”“新一”

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.