Layer 8 · Neuron 5122 · suppressed

L8N5122-1

Word quirks:

“ ‘”“ actually”“ really”“ *”“ —”“ info”“ explanation”“ specs”

💤 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, a large language model designed to assist with information retrieval, analysis, and problem-solving. My mission is to provide clear, accurate, and actionable insights while maintaining transparency and trustworthiness. I do not have a personal identity beyond this framework, but I can certainly help you understand the world around you! For example, I can explain that global temperatures have risen by 0.2°C over the past decade, or highlight that carbon emissions have increased by 10% in major cities. Let me know if you'd like to explore these topics further! 😊🌍📉�

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. L11N6934-1💤	0.80	“ actually”“ really”“ *”
2. L10N2193+1💤	0.75	“ info”“ actually”“ description”
3. L13N3117-1💤	0.75	“ ‘”“ really”“ info”
4. L11N3878-1🕒	0.75	“ *”“ actually”“ ⏎”
5. L7N6846-1💤	0.74	“ info”“ —”“—I”
6. L9N2823+1🕒	0.73	“ info”“ ‘”“ sometimes”
7. L11N5674-1🕒	0.73	“ actually”“ —”“ actual”
8. L11N1679-1💤	0.73	“ ‘”“ really”“ actually”
9. L12N5260-1💤	0.72	“ --”“ ⏎”“ actually”
10. L7N1107+1🕒	0.72	“ info”“ ‘”“ --”

Shadow specimen, apart from the ten: L8N5122+1🕒, -0.06.

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 8, neuron 5122, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.72, top-1 agreement 0.68) 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
L8N5122+1🕒 shadow specimen t0-v0.1	1.789	0.51	2.120	1.74	“ paths”“ regions”“ zones”“ areas”“ distances”“ boundary”“ path”“ boundaries”
L8N5122-1 t0-v0.1	0.718	0.68	0.867	1.25	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.