Layer 20 · Neuron 5529 · boosted

L20N5529+1

Word quirks:

“ destroying”“ destruction”“destroy”“ destroy”“Destroy”“ destroys”“ destroyed”“毁”

🕒 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 highly intelligent large language model developed by Alibaba Group. My primary purpose is to assist in breaking down complex problems, handling knowledge-based tasks, and providing creative support. I can help with:

Knowledge & Skills: Answering questions, writing content (e.g., articles, scripts), generating code, and analyzing data.
Tools & Efficiency: Breaking down complex tasks, handling multi-step operations, and optimizing workflows.
Creative Support: Assisting in brainstorming, storytelling, or creative design.
Language Proficiency: Supporting over

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. L25N2187+1🕒	0.99	“ Destroy”“destroy”“ destruction”
2. L31N7043-1💥	0.99	“ destruction”“ destroy”“destroy”
3. L20N5991-1🕒	0.99	“ destroying”“ destruction”“ destroy”
4. L25N4794+1🕒	0.98	“毁了”“ destroyed”“ Destruction”
5. L26N776-1🕒	0.98	“ destruction”“ destroy”“ Destruction”
6. L25N8605-1🕒	0.97	“ destruction”“破坏”“ destroy”
7. L1N5652-1💥	0.97	“ destroy”“ Destroy”“ destruction”
8. L2N7332+1💥	0.97	“ destruction”“<\|im_end\|>”“ destroy”
9. L23N1502+1🕒	0.97	“ Destruction”“ destruction”“ Destroy”
10. L28N6730-1💥	0.96	“破坏”“ destruction”“ destroy”

Shadow specimen, apart from the ten: L20N5529-1🕒, 0.07.

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 5529, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, top-1 agreement 0.63) 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
L20N5529+1 t0-v0.1	0.960	0.63	1.098	1.39	Top boosted tokens ↑
L20N5529-1🕒 shadow specimen t0-v0.1	0.684	0.66	0.765	1.26	“ align”“ Align”“ aligned”“ alignment”“ alignments”“对齐”“-&”“ transparent”

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.