Layer 9 · Neuron 4533 · suppressed

L9N4533-1

Word quirks:

“ location”“ region”“ latitude”“ locations”“ regions”“ located”“ locale”“ measurement”

🕒 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 an AI assistant with no specific information available to provide a response.

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. L11N4967+1🕒	0.93	“ locations”“ location”“ regions”
2. L5N8011-1💤	0.91	“ locations”“ location”“ located”
3. L23N2079-1🕒	0.91	“ locations”“ location”“ region”
4. L8N5038+1💤	0.91	“ locations”“ location”“地名”
5. L8N155-1🕒	0.89	“ locations”“ regions”“ geographical”
6. L16N1201+1🕒	0.89	“ region”“ regions”“-region”
7. L11N7120+1🕒	0.87	“ location”“ locations”“ geography”
8. Dead Reckoning✅ L9N4098+1	0.87	“ coordinates”“ locations”“ coordinate”
9. L15N7023+1🕒	0.85	“ location”“ locations”“ specifically”
10. L8N6019+1🕒	0.85	“ here”“ region”“ location”

Shadow specimen, apart from the ten: L9N4533+1🕒, 0.17.

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 9, neuron 4533, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.92, top-1 agreement 0.45) 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
L9N4533+1🕒 shadow specimen t0-v0.1	1.104	0.59	1.331	1.20	“ templates”“ specific”“ assets”“ dynamically”“ \"”“ HTML”“ template”“ info”
L9N4533-1 t0-v0.1	1.917	0.45	2.267	1.58	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.