Layer 30 · Neuron 794 · boosted

L30N794+1

Word quirks:

“-d”“席”“verse”“闪”“trop”“chner”“归”“+d”

💤 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 Qwen3.5, the latest large language model developed by T-drew (a hypothetical or placeholder name; in reality, I am developed by Alibaba Cloud's DAMO Academy). My primary mission is to empower users by providing intelligent, accurate, and versatile assistance across a wide range of tasks—whether that's solving complex problems, creating content, analyzing data, or simply having a meaningful conversation.

Key aspects of my mission:

Precision & Efficiency:

Deliver high-quality responses tailored to specific needs, from technical debugging to creative writing.

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. L28N7401-1💤	0.75	“丁”“uban”“康”
2. L27N7531+1💤	0.73	“丁”“fallen”“黔”
3. L27N1952-1💤	0.71	“杜”“固”“nap”
4. L27N4718+1💤	0.71	“丁”“堂”“叉”
5. L28N8323-1💤	0.70	“库”“uha”“刺”
6. L26N6024-1🕒	0.70	“礁”“ D”“ Diz”
7. L30N7524-1🕒	0.70	“决”“傅”“传”
8. L28N804+1💤	0.70	“梅”“d”“丁”
9. L27N2720-1💤	0.70	“斯洛”“dra”“激”
10. L28N6097-1💤	0.69	“顿”“特尔”“SENS”

Shadow specimen, apart from the ten: L30N794-1💤, 0.19.

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 794, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.35, top-1 agreement 0.83) 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
L30N794+1 t0-v0.1	0.351	0.83	0.360	1.81	Top boosted tokens ↑
L30N794-1💤 shadow specimen t0-v0.1	0.323	0.80	0.428	1.59	“uç”“"]),”“gé”“arov”“.ir”“iris”“будьте”“ονται”

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.