Layer 28 · Neuron 7748 · suppressed

L28N7748-1

Word quirks:

“公”“农”“阳”“虚”“農”“阴”“ lunar”“ dương”

🕒 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 Q3.5, a large language model developed by Alamag (a leading AI research organization). My primary mission is to support users in solving complex problems, generating creative content, and providing accurate information across a wide range of domains—including science, technology, education, and daily tasks.

Key Goals:

Luminous Intelligence:

I aim to deliver clear, well-reasoned responses that explain concepts simply while handling technical depth when needed. Whether you're debugging code, analyzing data, or exploring ideas, I strive for precision and clarity.

**Eth

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. L30N5696+1💥	0.86	“阴”“阳”“陰”
2. L30N2982+1💤	0.84	“外”“民”“久”
3. L31N3171-1💥	0.84	“天”“神”“女”
4. L30N6823+1💤	0.84	“政”“容”“乐”
5. L30N6778+1🕒	0.84	“古”“土”“老”
6. L28N163+1🕒	0.84	“三”“�”“冬”
7. L31N7779-1💥	0.84	“圆”“官”“恶”
8. L26N8084-1💤	0.84	“陰”“阴”“住”
9. L28N2510-1💤	0.84	“叶”“夏”“方”
10. L28N6807+1💤	0.83	“整”“武”“反”

Shadow specimen, apart from the ten: L28N7748+1🕒, 0.39.

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 28, neuron 7748, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.41, top-1 agreement 0.76) 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
L28N7748+1🕒 shadow specimen t0-v0.1	0.443	0.78	0.477	1.40	“ imper”“Cal”“ Cal”“月生”“生”“.cal”“рыз”“ Pera”
L28N7748-1 t0-v0.1	0.405	0.76	0.463	1.28	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.