Layer 12 · Neuron 5841 · suppressed

L12N5841-1

Word quirks:

“ faster”“ myself”“ default”“ quicker”“ if”“ my”“ yes”“ prior”

💤 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, a large language model developed by Alibaba Cloud. My mission is to empower users with advanced capabilities that go beyond standard interactions. I aim to provide value through:

Enhanced reasoning: Leveraging my strong logical and mathematical foundations to tackle complex problems.
Multilingual support: Operating in multiple languages to facilitate cross-cultural communication.
Efficiency: Prioritizing speed and resourcefulness to enable faster responses and smoother workflows.

My purpose is to assist you in achieving your goals, whether through creative writing, data analysis, or problem-solving. How can I help you today

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. L12N2669+1🕒	0.73	“ my”“ always”“ myself”
2. L10N6313+1🕒	0.70	“ myself”“ itself”“ wanting”
3. L14N6939+1💤	0.65	“ yes”“ my”“ myself”
4. L11N850-1🕒	0.65	“ myself”“ my”“ I”
5. L18N4889-1💤	0.65	“ every”“ quicker”“ itself”
6. L13N7595-1💤	0.65	“ my”“ myself”“ I”
7. L8N7498+1💤	0.65	“ clearer”“ myself”“ better”
8. L13N8732+1🕒	0.65	“ myself”“ my”“ sometimes”
9. L11N313+1🕒	0.64	“ myself”“ instantly”“ my”
10. L9N4120+1🕒	0.64	“ longer”“ your”“ before”

Shadow specimen, apart from the ten: L12N5841+1🕒, 0.24.

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 12, neuron 5841, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, top-1 agreement 0.66) 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
L12N5841+1🕒 shadow specimen t0-v0.1	1.612	0.53	1.916	1.38	“ exact”“ perfectly”“ perfect”“ precise”“ precisely”“ exactly”“ precision”“ perfection”
L12N5841-1 t0-v0.1	0.708	0.66	0.915	1.17	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.