Layer 19 · Neuron 5582 · suppressed

L19N5582-1

Word quirks:

“ divided”“ divide”“ dividing”“ divides”“divide”“ divid”“ Divide”“ div”

🕒 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)

Hello! I am Qwen, an AI assistant developed by Microsoft (specifically, I am part of the Qwen family of models from Microsoft's Azure cloud services). My primary mission is to empower users by providing efficient, creative, and intelligent assistance across a wide range of tasks.

Key Aspects of My Mission:

Problem Solving:

Break down complex problems into manageable steps for both technical and non-technical users.
Provide actionable solutions in fields like science, engineering, finance, or healthcare.

Creativity & Innovation:

Generate ideas

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. L19N2157-1💤	0.99	“ dividing”“ divide”“ divides”
2. L26N321+1🕒	0.99	“ divide”“ divided”“ division”
3. L25N458-1🕒	0.99	“ divide”“ division”“ Divide”
4. L20N6895+1🕒	0.98	“ divided”“ divide”“ dividing”
5. L23N3782+1🕒	0.98	“ divide”“ divided”“ divisions”
6. L23N7146+1🕒	0.97	“ divide”“ division”“ divides”
7. L20N8649-1🕒	0.97	“ divided”“除以”“ dividing”
8. L18N5379+1🕒	0.97	“ divides”“ divide”“ dividing”
9. L15N4545+1🕒	0.97	“ divided”“ dividing”“ divide”
10. L20N1566+1🕒	0.96	“ dividing”“ divides”“ anywhere”

Shadow specimen, apart from the ten: L19N5582+1💤, -0.07.

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 19, neuron 5582, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.69, top-1 agreement 0.70) 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
L19N5582+1💤 shadow specimen t0-v0.1	0.548	0.74	0.715	1.10	“ realization”“ParameterValue”“-word”“فضل”“ words”“ realism”“领悟”“ пости”
L19N5582-1 t0-v0.1	0.694	0.70	0.752	1.29	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.