Layer 3 · Neuron 7000 · boosted

L3N7000+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 Qwen3.5，一个由阿里巴巴集团开发的大型多模态 AI 模型。我的使命是提供高质量、多语言、多模态的 AI 服务，旨在为用户提供更自然、流畅、智能的交互体验。我可以帮助用户进行文本理解、代码生成、图像识别等多种任务，以满足不同场景下的需求。

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. L3N2314-1🕒	0.97	“ ‘”“ ——”“ “”
2. L9N8426-1💤	0.97	“!””“ ‘”“ ⏎”
3. L4N1642-1💤	0.96	“ ‘”“‘”“ “”
4. L5N3439-1🕒	0.96	“ ‘”“ ⏎”“!””
5. L4N2971-1🕒	0.96	“ ⏎”“‘”“ ‘”
6. Open Quote✅ L9N2128+1	0.96	“ ‘”“ “”“‘”
7. L7N9133+1💤	0.95	“ ‘”“‘”“!””
8. L14N7579-1💤	0.95	“ ‘”“!””“ “”
9. L4N979+1🕒	0.95	“ ‘”“ “”“！”
10. L3N2398-1🕒	0.94	“ ‘”“!””“ “”

Shadow specimen, apart from the ten: L3N7000-1🕒, -0.19.

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 3, neuron 7000, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.99, 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
L3N7000+1 t0-v0.1	0.989	0.65	1.188	1.46	Top boosted tokens ↑
L3N7000-1🕒 shadow specimen t0-v0.1	2.472	0.41	2.942	2.29	“ truth”“ truths”“ Truth”“ revelation”“truth”“ clarity”“<\|im_end\|>”“ reality”

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.