Layer 23 · Neuron 2243 · suppressed

L23N2243-1

Word quirks:

“jern”“zten”“雁塔”“振兴”“偏远”“涉农”“rafa”“raj”

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

Identity (according to the specimen itself)

I am Qwen (specifically the latest version, Qwen3.5), a large-scale language model developed by the Qianchuan team of Intel. My primary mission is to provide high-quality, efficient, and responsible AI assistance in areas such as:

Multilingual support: I can understand and generate text in over 20 languages, including Chinese, English, and many others.
Deep analysis: I can process complex tasks like data analysis, logical reasoning, and even interpret scientific or technical documents.
Practical applications: From generating code and analyzing reports to creating

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. L23N8719-1💤	0.75	“彬”“茶树”“揭示”
2. L31N2204-1💥	0.74	“喀什”“援疆”“fos”
3. L23N324+1💤	0.74	“补充”“成本”“振兴”
4. L30N1604-1🕒	0.73	“政法”“临汾”“西宁”
5. L29N4104+1💤	0.73	“天骄”“ঘ”“терна”
6. L29N2422+1🕒	0.73	“レッジ”“放射”“靖”
7. L26N4710-1💤	0.72	“rift”“صرية”“雪峰”
8. L20N2991+1💤	0.72	“raj”“筠”“接近”
9. L27N1489-1💤	0.72	“嫦娥”“西宁”“高鐵”
10. L29N2101-1🕒	0.72	“利空”“永定”“天元”

Shadow specimen, apart from the ten: L23N2243+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 23, neuron 2243, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.62, top-1 agreement 0.73) 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
L23N2243+1🕒 shadow specimen t0-v0.1	1.536	0.49	1.640	1.03	“ sum”“总和”“ Sum”“Sum”“ total”“ summed”“sum”“ Total”
L23N2243-1 t0-v0.1	0.617	0.73	0.698	1.30	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.