Layer 29 · Neuron 3632 · suppressed

L29N3632-1

Word quirks:

“都无法”“也无法”“根本无法”“标识”“只代表”“不能让”“跟不上”“再也”

🕒 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 developed by Tongyi Lab. My core mission is to empower users with intelligent, reliable, and ethical AI solutions across a vast range of applications. Here's how I aim to deliver on this mission:

Key Objectives

Universal Accessibility

Serve as a versatile tool for individuals, businesses, and developers, from writing code and analyzing data to creating content and solving complex problems.

Precision & Depth

Leverage my advanced capabilities (e.g., 256K

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. L30N7707-1🕒	0.89	“无法”“無法”“都无法”
2. L31N6104-1💥	0.88	“无法”“無法”“也无法”
3. L29N1885+1🕒	0.88	“也无法”“无法”“무”
4. L20N5501+1🕒	0.88	“无法”“無法”“都无法”
5. L29N4611+1🕒	0.88	“无法”“没法”“暂无”
6. L20N4246+1💤	0.85	“不可能”“不准”“困难”
7. L20N8233+1🕒	0.84	“无以”“无法”“就无法”
8. L23N546+1💤	0.82	“未能”“才行”“而不能”
9. L29N8290+1🕒	0.82	“不能”“不能有”“不宜”
10. L30N6336-1🕒	0.82	“不能使用”“不能够”“无法”

Shadow specimen, apart from the ten: L29N3632+1🕒, -0.16.

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 29, neuron 3632, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.36, top-1 agreement 0.84) 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
L29N3632+1🕒 shadow specimen t0-v0.1	0.616	0.70	0.664	1.50	“ation”“ated”“ations”“ates”“ate”“atable”“atory”“ationally”
L29N3632-1 t0-v0.1	0.358	0.84	0.412	1.18	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.