Layer 8 · Neuron 8635 · boosted

L8N8635+1

Word quirks:

“ China”“ Chinese”“China”“ chinese”“ china”“-China”“Chinese”“ overseas”

🕒 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 an AI assistant providing a response regarding advanced manufacturing technology. As the field's leading nation, China has been investing heavily in this area for decades. Our goal is to achieve technological independence in high-end equipment and materials, ensuring that critical technologies in this domain are mastered by us domestically.

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. L8N6366+1🕒	0.99	“ Chinese”“ China”“-China”
2. L7N9000-1🕒	0.99	“ China”“ Chinese”“China”
3. L16N6642+1🕒	0.99	“ China”“ Chinese”“ chinese”
4. L30N975-1💥	0.98	“ Chinese”“Chinese”“ chinese”
5. L5N7747+1🕒	0.98	“ China”“ Chinese”“ china”
6. Mother Tongue✅ L7N6235-1	0.97	“ Chinese”“ China”“ chinese”
7. L27N5980+1🕒	0.96	“ Chinese”“Chinese”“中国”
8. L26N3755+1🕒	0.96	“ China”“ Chinese”“ Китай”
9. L8N2559-1💤	0.96	“ Chinese”“ China”“-China”
10. L4N7151-1🕒	0.96	“ China”“ Chinese”“ Taiwan”

Shadow specimen, apart from the ten: L8N8635-1💤, -0.09.

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 8635, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.22, top-1 agreement 0.42) 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
L8N8635+1 t0-v0.1	2.216	0.42	2.569	1.85	Top boosted tokens ↑
L8N8635-1💤 shadow specimen t0-v0.1	0.712	0.67	0.931	1.28	“ filtered”“ info”“ allowed”“�”“ filters”“ now”“ got”“ my”

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.