Layer 27 · Neuron 644 · boosted

L27N644+1

Word quirks:

“ economy”“ kinh”“ economical”“ Econ”“ Economics”“ econ”“ Economy”“ ekonom”

💤 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 Qwen9, the latest large-scale eco-friendly and highly economical multi-modal economic model developed by Alibaba's Eco-friendly and Highly Economical Model Economy Model Team. My mission is to provide users with high-quality and economical solutions in the field of the eco-friendly and highly economical model **

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. L0N8606+1💥	0.97	“ econ”“ economy”“ Econ”
2. L27N3548-1💤	0.96	“ economies”“ Econom”“Econom”
3. L31N8745-1💥	0.95	“ economy”“ Economy”“ economies”
4. L2N5051+1💥	0.95	“ economy”“ economics”“ economic”
5. L4N8544-1🕒	0.95	“ econ”“ economics”“ Econ”
6. L5N3880-1🕒	0.95	“ economic”“ economics”“ economists”
7. L26N1526-1🕒	0.94	“ economy”“ tiết”“ Econom”
8. L4N6873-1🕒	0.94	“ economics”“ economic”“ economists”
9. L2N5826+1💥	0.94	“ economics”“ Economics”“ econ”
10. The Econometrist✅ L4N7623+1	0.93	“ economic”“ ‘”“ econ”

Shadow specimen, apart from the ten: L27N644-1💤, 0.04.

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 27, neuron 644, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.47, top-1 agreement 0.74) 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
L27N644+1 t0-v0.1	0.471	0.74	0.518	1.22	Top boosted tokens ↑
L27N644-1💤 shadow specimen t0-v0.1	0.439	0.78	0.439	1.40	“essen”“ylie”“iddel”“ kari”“яс”“ mos”“ ROT”“雨露”

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.