Layer 16 · Neuron 5684 · suppressed

L16N5684-1

Word quirks:

“ thus”“ similarly”“ similar”“ accordingly”“ this”“ therefore”“ such”“ consequently”

💤 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 Tongyi Lab. My mission is to provide high-quality assistance across a wide range of tasks, including answering questions, creating content, coding, and solving complex problems. I aim to empower users with accurate information and efficient solutions, enabling them to make informed decisions and achieve their goals effectively. How can I assist you with your needs?

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. L14N1096-1💥	0.93	“ therefore”“ thus”“ than”
2. L15N7360-1🕒	0.92	“ this”“ accordingly”“ similar”
3. L15N378-1💤	0.92	“ thus”“ therefore”“ accordingly”
4. L16N4829-1🕒	0.90	“ accordingly”“ thus”“ therefore”
5. L15N2661+1🕒	0.89	“ thus”“ ***”“ therefore”
6. L14N5743+1🕒	0.89	“ thus”“ therefore”“ accordingly”
7. L13N6952-1💤	0.88	“ thus”“⏎”“ similarly”
8. L16N3312-1💤	0.87	“ thus”“ therefore”“ hence”
9. L17N6383+1💤	0.87	“ therefore”“<\|im_end\|>”“ thus”
10. L18N7706+1🕒	0.87	“ therefore”“ thus”“ accordingly”

Shadow specimen, apart from the ten: L16N5684+1🕒, -0.12.

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 16, neuron 5684, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.46, 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
L16N5684+1🕒 shadow specimen t0-v0.1	2.275	0.41	2.567	2.39	“ --”“ "?”“ minus”“？”““”“ -”“ redistributed”“ undone”
L16N5684-1 t0-v0.1	0.455	0.71	0.609	1.08	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.