Layer 20 · Neuron 5665 · boosted

L20N5665+1

Word quirks:

“标准”“ Standard”“-standard”“Standard”“稳固”“标准的”“ dur”“固”

💤 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 QZAE, an advanced AI system designed to assist users with a wide range of tasks, from creative problem-solving to factual queries. My mission is to provide clear, thoughtful, and helpful responses that align with the user's needs while adhering to safety guidelines. I aim to be a reliable partner for learning, creativity, and decision-making, always striving to deliver value without ambiguity or harm. How would you like to start?

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. L26N7781-1🕒	0.91	“ Standard”“标准”“的标准”
2. L25N2650-1🕒	0.91	“ standard”“标准”“ Standard”
3. L20N8700-1🕒	0.90	“ standard”“标准”“standard”
4. L20N7040+1🕒	0.90	“ standard”“标准”“標準”
5. L26N4432-1🕒	0.90	“ Standard”“标准和”“标准”
6. L29N5654-1🕒	0.90	“ standard”“标准”“standard”
7. L31N952-1💥	0.90	“标准”“ standard”“ Standard”
8. L31N8068-1💥	0.90	“标准”“标准要求”“Standard”
9. L25N3672+1🕒	0.90	“ Standard”“标准化”“标准”
10. L24N4159-1💤	0.90	“孔”“标准”“ Standard”

Shadow specimen, apart from the ten: L20N5665-1💤, -0.02.

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 20, neuron 5665, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.62, top-1 agreement 0.69) 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
L20N5665+1 t0-v0.1	0.616	0.69	0.759	1.29	Top boosted tokens ↑
L20N5665-1💤 shadow specimen t0-v0.1	0.578	0.73	0.649	1.26	“ phys”“ predecessors”“ Phys”“ predecessor”“iston”“ fisica”“imon”“ when”

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.