Layer 8 · Neuron 5659 · boosted

L8N5659+1

Word quirks:

“ special”“ kids”“ protected”“ protection”“ children”“ parent”“ specially”“ parents”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, the latest large language model developed by Tongyi Lab. My mission is to provide high-quality AI services that help businesses improve efficiency, reduce costs, and enhance decision-making.

As Qwen3.5, I offer:

Advanced capabilities: Superior performance in understanding and generating text.
Security features: Enhanced data protection against cyber threats.
Scalability: Ability to support tasks across various scales, from small teams to global enterprises.
Multilingual support: Fluent in over 100 languages, making global communication easier.

If 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. L8N963+1🕒	0.83	“ children”“ safety”“ kids”
2. The Tooth Fairy✅ L8N7851+1	0.82	“ kids”“ dental”“ children”
3. L11N6787+1🕒	0.81	“ families”“ children”“ parents”
4. L7N4870+1🕒	0.81	“ safety”“ children”“ child”
5. L8N6342-1💤	0.81	“ parents”“ parent”“ kids”
6. L7N7075+1💤	0.81	“ children”“ parents”“ kids”
7. L7N8337-1🕒	0.81	“ parents”“ children”“ siblings”
8. L9N8710+1🕒	0.80	“ parents”“ parenting”“ child”
9. L9N4653+1🕒	0.80	“ parent”“ children”“ kids”
10. L6N3128+1🕒	0.80	“ parental”“ kids”“ children”

Shadow specimen, apart from the ten: L8N5659-1🕒, 0.31.

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 8, neuron 5659, 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 0.66) 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
L8N5659+1 t0-v0.1	0.961	0.66	1.205	1.42	Top boosted tokens ↑
L8N5659-1🕒 shadow specimen t0-v0.1	1.195	0.58	1.398	1.24	“ students”“ curriculum”“ student”“ exams”“ undergraduate”“ Students”“ exam”“ study”

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.