Layer 30 · Neuron 650 · boosted

L30N650+1

Word quirks:

“ußen”“ñado”“JKLMNOP”“zwi”“�”“amik”“ tille”“里斯”

🕒 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 Qwen3.5, the latest large language model developed by Tuguo AI (Tuguo). My primary mission is to empower human potential by providing intelligent, accurate, and ethically grounded assistance across a wide range of domains.

Key Objectives:

Solve Complex Problems:

Tackle multi-step reasoning, mathematical derivations, and scientific analysis with precision.
Generate high-quality code, debug errors, and optimize software architectures.

Enhance Human-Centric AI:

Deliver natural, context-aware responses in **1

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. L29N1866+1💤	0.69	“аниз”“yal”“uus”
2. L31N1929-1💥	0.69	“ilos”“�”“ampus”
3. L31N2006+1💥	0.68	“ämm”“estero”“ugio”
4. L30N549+1💤	0.68	“ظم”“-Za”“wayne”
5. L30N1878+1💤	0.67	“يمي”“риз”“ím”
6. L31N526+1💥	0.67	“ cuckold”“Gatt”“vrier”
7. L30N1736+1🕒	0.67	“jus”“uzzi”“adius”
8. L30N6263-1💤	0.67	“MU”“adá”“freund”
9. L31N3062+1💥	0.67	“adolid”“ushman”“ailles”
10. L31N7623+1💥	0.66	“藏”“čit”“رائ”

Shadow specimen, apart from the ten: L30N650-1💤, 0.44.

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 30, neuron 650, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.81) 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
L30N650+1 t0-v0.1	0.697	0.81	1.008	3.24	Top boosted tokens ↑
L30N650-1💤 shadow specimen t0-v0.1	0.166	0.86	0.194	0.89	“istus”“เทศ”“agement”“arra”“sing”“parser”“dbuf”“ourd”

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.