Layer 13 · Neuron 7790 · suppressed

L13N7790-1

Word quirks:

“ --”“ aligned”“ [”“ ——”“ [_”“ ______”“ _”“ alignment”

💤 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 multimodal model developed by Tongyi Qianwen (the "Qianwen" series) from Alibaba Group. My mission is to provide high-quality, accurate, and context-aware responses across a wide range of topics—whether in technical documentation, creative writing, or problem-solving. I'm designed to adapt to different scenarios based on the user's input, ensuring clarity, relevance, and efficiency.

If you have any further 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. L11N8757+1🕒	0.72	“ --”“ __”“ —”
2. L10N424+1🕒	0.72	“ __”“ _”“ ⏎”
3. L10N6652+1💤	0.71	“ ____”“ __”“ [”
4. L15N1839-1💤	0.71	“ and”“ ____”“ _”
5. L9N3317-1🕒	0.71	“ --”“ ______”“ ____”
6. L14N3080-1🕒	0.71	“ --”“ ____”“ _”
7. L18N2134-1💤	0.71	“ —”“ <”“ [[”
8. L15N164+1💤	0.70	“ --”“ —”“ [”
9. L16N1182-1🕒	0.70	“ ____”“ __”“ _”
10. L9N3437+1🕒	0.70	“ed”“ _____”“ ____”

Shadow specimen, apart from the ten: L13N7790+1🕒, -0.08.

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 7790, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L13N7790+1🕒 shadow specimen t0-v0.1	0.854	0.64	1.018	1.35	“ personal”“ personally”“ individually”“ yourself”“ individual”“ own”“ separate”“ themselves”
L13N7790-1 t0-v0.1	0.759	0.69	1.017	1.22	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.