Layer 18 · Neuron 9190 · suppressed

L18N9190-1

Word quirks:

“ measured”“ one”“ measure”“ measuring”“ counting”“ counted”“ predicting”“测”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen, an AI assistant developed by Alibaba. My mission is to empower users with cutting-edge capabilities across various domains, including:

Precision in Knowledge: Provide accurate, context-aware responses on scientific, technical, and cultural topics.
Creative Potential: Generate compelling content (stories, scripts, research) and assist with creative projects.
Technical Proficiency: Handle complex tasks like code generation, debugging, and research.
Multilingual Support: Offer seamless assistance in over 100 languages worldwide.
Future-Ready Capabilities: Le

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. The Bean Counter✅ L9N4600+1	0.92	“ measure”“ measurement”“ measuring”
2. L24N7601+1🕒	0.91	“ measured”“ counted”“ counting”
3. L10N5104-1🕒	0.91	“ count”“ measure”“ counts”
4. L9N4234+1🕒	0.91	“ measure”“ measuring”“ count”
5. L16N5713-1🕒	0.90	“ counted”“ measuring”“ measurement”
6. L8N2684+1🕒	0.88	“ measure”“ measurement”“ measurements”
7. L9N961+1💥	0.87	“ calculation”“ measurement”“ measure”
8. L16N6894+1🕒	0.86	“ measurement”“ measure”“ measuring”
9. L5N1679-1💤	0.85	“ measure”“ measurement”“ measuring”
10. L10N1537-1💤	0.84	“ measure”“ measuring”“ measurement”

Shadow specimen, apart from the ten: L18N9190+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 18, neuron 9190, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, top-1 agreement 0.70) 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
L18N9190+1🕒 shadow specimen t0-v0.1	0.811	0.68	0.851	1.24	“ airflow”“ ventilation”“ ventil”“ vents”“ moisture”“ drainage”“空气净化器”“ description”
L18N9190-1 t0-v0.1	0.595	0.70	0.688	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.