Layer 12 · Neuron 3760 · suppressed

L12N3760-1

Word quirks:

“⏎”“ **”“ **“”“...**”“ **"”“ **'”“ **-”“ ***”

💤 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, the latest large language model in the Qwen series. I am designed to handle a wide range of tasks with enhanced capabilities. My training data is based on a more comprehensive knowledge corpus, allowing me to understand and generate content with higher quality and accuracy.

I am part of the Qwen3.5 family, which builds upon the previous version (Qwen3). This iteration further improves my performance in areas like logical reasoning, mathematical problem-solving, and code generation. I support more complex multi-step tasks, including advanced analysis and reasoning.

In summary:

I am

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. L14N7771-1🕒	0.96	“ "”“ ___”“ '”
2. L14N3242-1💤	0.95	“ ”“ "”“...**”
3. The Footnote Machine✅ L16N7785+1	0.95	“ *”“ ”“ **'”
4. L10N2425+1💤	0.95	“ "”“ ”“⏎”
5. L15N3784+1💤	0.95	“ "”“ “”“ "**”
6. L16N3048-1💤	0.95	“ “”“...”“ **[”
7. L8N120-1💤	0.95	“ *”“ ”“ NOT”
8. L16N3637-1💤	0.95	“ “”“ ⏎”“ ”
9. L8N937-1💤	0.95	“ ”“ “”“ **‘”
10. L15N3500+1🕒	0.95	“ "”“ ”“...**”

Shadow specimen, apart from the ten: L12N3760+1💥, 0.34.

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 3760, 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 0.71) 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
L12N3760+1💥 shadow specimen t0-v0.1	7.143	0.09	7.633	3.44	“"@”“ "@”“ "_”“"”“."_”“"_”“":”“lett”
L12N3760-1 t0-v0.1	0.586	0.71	0.745	1.23	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.