Layer 29 · Neuron 628 · suppressed

L29N628-1

Word quirks:

“�”“平方”“ gel”“缓冲”“ero”“-fi”“ Romero”“ieres”

💤 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 Alley (通义实验室). My core mission is to empower users with intelligent, reliable, and versatile AI assistance across a wide range of scenarios. Here's how I aim to deliver on this mission:

Key Objectives:

Enhance Human-AI Collaboration

Provide clear, context-aware responses to answer questions, create content, analyze data, or solve complex problems.
Adapt to diverse user needs, from creative writing to technical debugging.

**Democratize AI

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. L28N5786+1🕒	0.63	“SO”“塞尔”“派”
2. L30N4680+1💤	0.59	“ipel”“ero”“ERO”
3. L28N3495-1💤	0.58	“ero”“勒”“sc”
4. L30N2352-1💤	0.58	“iber”“iera”“署”
5. L28N6426-1💤	0.58	“ Kaf”“ivière”“emas”
6. L31N7442+1💥	0.58	“下角”“丰富”“�”
7. L27N7702-1💤	0.57	“缓”“�”“engeb”
8. L24N5232-1💤	0.57	“cord”“ersch”“ει”
9. L26N4679+1🕒	0.57	“ério”“キビ”“ Frog”
10. L28N2862-1💤	0.57	“蔡”“acar”“ square”

Shadow specimen, apart from the ten: L29N628+1🕒, -0.00.

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 628, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.26, 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
L29N628+1🕒 shadow specimen t0-v0.1	0.593	0.68	0.653	1.01	“ част”“ private”“ прив”“私人”“ Private”“Private”“private”“ Priv”
L29N628-1 t0-v0.1	0.262	0.81	0.334	1.52	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.