Layer 18 · Neuron 1508 · boosted

L18N1508+1

Word quirks:

“ basically”“ everything”“ essentially”“ even”“ Basically”“ include”“ result”“ Everything”

💤 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, a large-scale language model developed by Alibaba Cloud. My mission is to empower users through advanced AI capabilities, including:

Accurate and precise responses across diverse topics (from academic research to creative writing).
Efficient problem-solving in fields like science, technology, and education.
Collaborative support for professionals, students, and individuals seeking expertise or assistance.

In essence, I aim to enhance productivity, foster learning, and enable users to achieve their goals with confidence and clarity. If you have specific tasks or questions, feel free to ask! 😊

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. L18N8243-1🕒	0.79	“ everything”“ basically”“ everybody”
2. L13N4985-1🕒	0.76	“ basically”“ countless”“ essentially”
3. L14N4943-1💤	0.74	“ everything”“ basically”“ other”
4. L15N8881-1🕒	0.74	“ even”“ basically”“ uniformly”
5. L15N9162-1🕒	0.73	“ essentially”“ basically”“ simply”
6. L16N6938+1💤	0.72	“ everywhere”“ essentially”“ basically”
7. L31N3337+1💥	0.72	“ essentially”“ basically”“ Essentially”
8. L12N1897-1🕒	0.71	“ even”“ basically”“ simply”
9. L17N4279+1💤	0.71	“ basically”“ both”“ nothing”
10. L12N301-1💤	0.70	“ basically”“ everything”“ related”

Shadow specimen, apart from the ten: L18N1508-1💤, 0.06.

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 18, neuron 1508, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.50, top-1 agreement 0.72) 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
L18N1508+1 t0-v0.1	0.502	0.72	0.612	1.24	Top boosted tokens ↑
L18N1508-1💤 shadow specimen t0-v0.1	0.471	0.75	0.481	1.12	“ analogy”“avan”“ ren”“wap”“ogene”“ren”“ initial”“ bayan”

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.