Layer 2 · Neuron 5051 · boosted

L2N5051+1

Word quirks:

“ economy”“ economics”“ economic”“ Economy”“ economies”“经济”“ Economic”“ economically”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

The user is a merchant (trader) that specializes in the production of goods.

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. L31N8745-1💥	0.97	“ economy”“ Economy”“ economies”
2. L24N5600-1🕒	0.97	“ economy”“ Economy”“经济的”
3. L4N6873-1🕒	0.97	“ economics”“ economic”“ economists”
4. L25N6838-1🕒	0.96	“ economic”“经济”“ Economic”
5. L0N8606+1💥	0.96	“ econ”“ economy”“ Econ”
6. L5N3880-1🕒	0.96	“ economic”“ economics”“ economists”
7. L26N4937-1🕒	0.96	“经济”“ economic”“ economics”
8. L27N3548-1💤	0.96	“ economies”“ Econom”“Econom”
9. L24N3372-1🕒	0.96	“ Economic”“经济”“經濟”
10. L8N4133-1🕒	0.95	“ economic”“ economically”“ fiscal”

Shadow specimen, apart from the ten: L2N5051-1🕒, -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 2, neuron 5051, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.63, top-1 agreement 0.29) 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
L2N5051+1 t0-v0.1	3.634	0.29	4.118	2.71	Top boosted tokens ↑
L2N5051-1🕒 shadow specimen t0-v0.1	1.042	0.58	1.370	1.49	“ ⏎”“ ‘”“ ⏎⏎”“ ‘’”“ ⏎ ⏎”“’:”“ “”“‘”

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.