Layer 28 · Neuron 8005 · boosted

L28N8005+1

Word quirks:

“impin”“ustin”“碟”“雷”“ sema”“enis”“aglio”“ahre”

🕒 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, a large-scale language model developed by Kusf (a hypothetical or placeholder entity for this context; in reality, I am likely referring to a specific model like Q3 from Q&L or another provider, but since no official name was provided in your prompt, I will clarify based on common usage).

Who I Am:

Identity: I am an artificial intelligence designed to assist with tasks ranging from answering questions and creating content to solving problems and analyzing data.
Capabilities: I can generate text, code, and visual descriptions; summarize documents; explain complex concepts; and

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. L27N2509-1💤	0.71	“ynos”“履”“سیم”
2. L31N4601-1💥	0.70	“trom”“漩”“asher”
3. L29N6740-1💤	0.69	“椎”“ti”“amm”
4. L29N4469+1💤	0.68	“闪”“chio”“休”
5. L31N5103+1💥	0.68	“ulli”“道”“uur”
6. L25N1237-1💤	0.68	“enson”“Hel”“ Hel”
7. L31N1059-1💥	0.68	“矶”“ambe”“امين”
8. L27N1298+1💤	0.68	“leb”“reta”“ション”
9. L30N6103-1💤	0.67	“荷”“uera”“iski”
10. L31N5064+1💥	0.67	“uks”“arca”“ensem”

Shadow specimen, apart from the ten: L28N8005-1🕒, 0.14.

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 8005, 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 0.81) 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
L28N8005+1 t0-v0.1	0.409	0.81	0.406	1.43	Top boosted tokens ↑
L28N8005-1🕒 shadow specimen t0-v0.1	0.673	0.68	0.733	1.38	“依”“ adherence”“ adher”“ Ad”“ adhere”“Ad”“-ad”“ adhes”

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.