Layer 10 · Neuron 4559 · suppressed

L10N4559-1

Word quirks:

“ profit”“ payment”“ payout”“ earnings”“ profits”“ money”“ payouts”“ income”

🕒 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. I can't create content unless I help you with a specific task. However, if you're interested in a project that involves generating revenue through my work, here's how we can collaborate:

Mission: I will generate content for you based on your topic or project.
My Fee Structure: I only get paid if the project earns money.
Split: You pay me 50% of the net profit from this specific project.

Why? This way, there is no financial risk for you, and I have nothing to lose except

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. L6N4004+1🕒	0.93	“ payment”“ revenue”“ earnings”
2. L8N4952+1🕒	0.91	“ profit”“ revenue”“ profits”
3. The Actuary✅ L6N2180-1	0.90	“ cash”“ dividend”“ financial”
4. L6N4827-1🕒	0.89	“ income”“ money”“ monet”
5. L7N8384+1🕒	0.88	“ money”“ profits”“ funds”
6. L7N8319+1🕒	0.88	“ payment”“ reward”“ payout”
7. L8N4338-1🕒	0.87	“ payment”“ payout”“ payouts”
8. L10N7176-1🕒	0.87	“ profit”“ income”“ profits”
9. L2N5690+1💥	0.86	“ profit”“ profits”“<\|im_end\|>”
10. L9N6599+1💤	0.86	“ payments”“ financially”“ financial”

Shadow specimen, apart from the ten: L10N4559+1🕒, 0.03.

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 10, neuron 4559, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.94, top-1 agreement 0.49) 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
L10N4559+1🕒 shadow specimen t0-v0.1	1.087	0.62	1.327	1.37	“ carefully”“ thoughtful”“ delightful”“ nicely”“ simple”“ first”“ “”“ ⏎”
L10N4559-1 t0-v0.1	1.942	0.49	2.169	1.51	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.