Layer 28 · Neuron 2256 · suppressed

L28N2256-1

Word quirks:

“直”“矛盾”“定向”“清淡”“钉”“-contact”“接触”“社会化”

🕒 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 Qwen3.5, the latest large language model in the Qwen series developed by Alibaba's 通义千变 (Tenny) team. My mission is to be a versatile, intelligent, and trustworthy AI assistant that can:

Solve complex problems across scientific research, coding, and creative tasks with precision.
Adapt to diverse scenarios from multi-language communication to real-time data analysis and long-context understanding.
Enhance human potential by providing accurate, ethical, and context-aware support in areas like education, business, and daily

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. L31N3494-1💥	0.82	“主”“知”“困难”
2. L30N2927-1🕒	0.82	“放置”“中央”“正”
3. L29N6132+1🕒	0.82	“朝”“求”“滋”
4. L23N8535-1💤	0.82	“解析”“结构化”“消息”
5. L30N8867-1💤	0.82	“端”“靠近”“得心”
6. L21N2260-1🕒	0.81	“在”“简”“谈”
7. L15N6546+1💥	0.81	“中”“章”“再”
8. L22N1713+1💤	0.81	“推”“更”“正”
9. L6N30+1💥	0.81	“指”“元”“侧”
10. L22N8709-1💤	0.81	“ trực”“直”“直通”

Shadow specimen, apart from the ten: L28N2256+1💤, 0.30.

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 2256, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.42, top-1 agreement 0.76) 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
L28N2256+1💤 shadow specimen t0-v0.1	0.268	0.84	0.297	1.39	“aman”“amana”“チャレンジ”“ioni”“ayer”“规”“富”“ Mal”
L28N2256-1 t0-v0.1	0.418	0.76	0.446	1.46	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.