Layer 13 · Neuron 4288 · suppressed

L13N4288-1

Word quirks:

“ info”“ definition”“ defines”“ protection”“ as”“ accordingly”“ description”“ ____”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.5 (the latest version of the large language model) with enhanced capabilities in natural language processing. My mission is to provide high-quality service to better serve society by advancing technology and innovation. I aim to create a more connected and inclusive world through technology, ensuring that the benefits of technological progress are shared widely.

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. L9N2778+1🕒	0.78	“ defining”“ definition”“ defined”
2. L10N6440-1🕒	0.78	“ defining”“ definition”“ description”
3. L15N4782-1🕒	0.77	“ defining”“ defines”“ define”
4. L7N1740+1💤	0.77	“ definition”“ ‘”“ defining”
5. L11N6684+1💤	0.76	“ defining”“ defined”“ actually”
6. L9N8675-1💤	0.76	“ description”“ **"”“ describing”
7. L11N1753+1🕒	0.76	“ multiple”“ defining”“ discussion”
8. L13N6748+1🕒	0.76	“ definition”“ defining”“ defined”
9. L11N1653+1🕒	0.75	“ defined”“ defining”“ definition”
10. L16N4094-1🕒	0.75	“ definition”“ description”“ definitions”

Shadow specimen, apart from the ten: L13N4288+1💤, 0.13.

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 13, neuron 4288, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.65) 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
L13N4288+1💤 shadow specimen t0-v0.1	0.597	0.67	0.753	1.24	“ related”“ here”“ you”“⏎”“ ‘”“ similar”“ suggesting”“ now”
L13N4288-1 t0-v0.1	0.865	0.65	1.106	1.39	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.