Layer 4 · Neuron 5285 · boosted

L4N5285+1

Word quirks:

“ compensation”“ money”“ payment”“ financial”“ legal”“ financially”“ funds”“ business”

🕒 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 designed to be helpful, harmless, and honest. My mission is to provide accurate, reliable, and ethical assistance to users who seek information, support, or guidance on a wide range of topics.

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. L8N3393-1🕒	0.92	“ money”“ financial”“ funds”
2. The Loan Officer✅ L8N4035+1	0.92	“ financial”“ payment”“ payments”
3. L7N2331+1🕒	0.89	“ business”“ monet”“ financially”
4. L9N6599+1💤	0.89	“ payments”“ financially”“ financial”
5. L7N8845+1🕒	0.89	“ growth”“ financing”“ financial”
6. L7N2627-1🕒	0.89	“ financial”“ investment”“ Financial”
7. L6N6679+1🕒	0.89	“ financial”“ fiscal”“ financially”
8. L5N4084-1💤	0.89	“ money”“ financial”“ cash”
9. L9N4304-1🕒	0.89	“ financial”“ money”“ funds”
10. L7N8384+1🕒	0.88	“ money”“ profits”“ funds”

Shadow specimen, apart from the ten: L4N5285-1💤, -0.10.

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 4, neuron 5285, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.06, top-1 agreement 0.47) 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
L4N5285+1 t0-v0.1	2.057	0.47	2.350	1.98	Top boosted tokens ↑
L4N5285-1💤 shadow specimen t0-v0.1	0.879	0.62	1.079	1.66	“ ‘”“‘”“ “”“ ‘”“ “”“�”“ ‘’”“⏎”

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.