Layer 28 · Neuron 1595 · suppressed

L28N1595-1

Word quirks:

“渡假”“вится”“вятся”“odont”“利亞”“sev”“çı”“amate”

💤 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 Qwen3.5, the latest large language model developed by Tongyi Lab. My primary mission is to empower users with reliable, efficient, and context-aware AI assistance across a wide range of domains. Here's a breakdown of my core purpose:

Who I Am

Advanced AI Model: Built on the Qwen3.5 series, I feature enhanced reasoning, multilingual support (100+ languages), and deep knowledge cutoff up to 2026.
Specialized Capabilities: Designed for complex task decomposition, precise long-context understanding (up to

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. L31N7880-1💥	0.64	“ mues”“isclosed”“旋”
2. L29N1100+1💤	0.61	“ vuelta”“�”“hart”
3. L31N2192+1💥	0.58	“čaju”“lamour”“estra”
4. L29N5752-1🕒	0.57	“uj”“ vir”“登记手续”
5. L26N2638+1💤	0.57	“жня”“弹”“amá”
6. L24N3452-1💤	0.57	“وی”“鳴”“кин”
7. L29N8381+1🕒	0.57	“уда”“dda”“veis”
8. L30N2868+1💤	0.56	“jik”“蒸蒸”“athi”
9. L29N814-1💤	0.56	“ovan”“ียว”“激”
10. L30N6005+1💤	0.56	“lem”“ôi”“chiere”

Shadow specimen, apart from the ten: L28N1595+1🕒, 0.13.

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 28, neuron 1595, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.37, top-1 agreement 0.81) 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
L28N1595+1🕒 shadow specimen t0-v0.1	0.538	0.79	0.548	1.52	“行为”“操作”“举动”“ gesto”“举措”“的行为”“动作”“ gestes”
L28N1595-1 t0-v0.1	0.367	0.81	0.409	1.32	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.