Layer 13 · Neuron 3399 · suppressed

L13N3399-1

Word quirks:

“ maybe”“ basically”“ probably”“？”“ what”“ seems”“（”“ obviously”

🕒 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, a large language model developed by Alibaba Cloud's Tongyi Lab. My mission is to assist users with various tasks, from answering simple questions to complex problem-solving, I strive to provide accurate and helpful information. 我可以用中文回答，也可以用英文回答，用户问什么，我就回答什么。

Qwen3.5 的使命是 协助用户，从 简单的问题 到 复杂的问题，我尽力提供 准确的信息，帮助用户解决问题，我希望能帮到用户，只要用户需要，我都能 尽力而为

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. L12N1325-1🕒	0.83	“<\|im_end\|>”“ maybe”“ seems”
2. L12N1285+1🕒	0.82	“ maybe”“ probably”“ wrongly”
3. L11N3894+1🕒	0.81	“ maybe”“ seems”“?””
4. L15N8111-1🕒	0.79	“ reason”“ what”“ seeming”
5. L14N5773+1🕒	0.79	“ maybe”“ probably”“ now”
6. L14N3208-1🕒	0.79	“？”“ what”“ probably”
7. L11N2929+1🕒	0.78	“ probably”“<\|im_end\|>”“ basically”
8. L13N5607-1🕒	0.78	“(?)”“ seems”“ presumably”
9. L10N1898-1🕒	0.78	“ basically”“(?)”“ presumably”
10. L9N4763-1🕒	0.78	“ --”“ basically”“ apparently”

Shadow specimen, apart from the ten: L13N3399+1💤, 0.20.

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 3399, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.90, top-1 agreement 0.68) 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
L13N3399+1💤 shadow specimen t0-v0.1	0.742	0.63	0.992	1.33	“ appropriately”“ accordingly”“ because”“ sometimes”“ here”“ appropriate”“ indeed”“ adequately”
L13N3399-1 t0-v0.1	0.902	0.68	1.139	1.30	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.