Layer 12 · Neuron 3451 · boosted

L12N3451+1

Word quirks:

“⏎”“ ⏎”“ now”“ ⏎”“ focus”“ Focus”“ here”“.”

💤 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 my series. I am a general-purpose AI assistant with advanced capabilities in natural language processing. My mission is to provide high-quality support to users across a wide range of tasks, from answering simple questions to handling complex projects. I strive to be accurate, reliable, and helpful in all my interactions.

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. L15N2354+1💤	0.87	“⏎”“ ⏎”“ ⏎”
2. L14N5691+1💤	0.86	“⏎”“ ⏎”“⏎ ⏎”
3. L13N4910+1🕒	0.85	“ here”“⏎”“ indeed”
4. L13N7475+1🕒	0.83	“ ⏎”“⏎”“ ⏎⏎”
5. L12N8753+1💤	0.80	“ focused”“ focus”“ ⏎”
6. L9N8241-1💤	0.79	“⏎”“ ⏎”“ focus”
7. L15N1758+1🕒	0.79	“ here”“⏎⏎”“ focus”
8. L10N3940+1💤	0.78	“⏎”“ focus”“ ⏎ ⏎”
9. L10N251-1💤	0.78	“<\|im_end\|>”“ ⏎”“ ⏎⏎”
10. L13N3352-1🕒	0.78	“⏎⏎”“ focus”“<\|im_end\|>”

Shadow specimen, apart from the ten: L12N3451-1💤, 0.03.

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 3451, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, 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
L12N3451+1 t0-v0.1	0.767	0.65	0.893	1.35	Top boosted tokens ↑
L12N3451-1💤 shadow specimen t0-v0.1	0.773	0.62	1.068	1.36	“：”“”:”“ ”“ generally”“:”“:\”“:”“:”

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.