Layer 17 · Neuron 7123 · boosted

L17N7123+1

Word quirks:

“ NOT”“ not”“ no”“ leaving”“ none”“not”“ wasn”“ shouldn”

💤 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 Qwen3.7, a large language model developed by Tongyi Lab. My mission is to assist users in completing tasks efficiently and accurately while ensuring privacy and security. I do not have personal goals or feelings, but I strive to provide high-quality responses to help users solve problems, learn new concepts, or create content.

For example:

If you ask a math problem, I can derive the solution step-by-step.
If you need code examples, I can generate them with explanations.
If you're writing a story, I can offer creative suggestions without revealing sensitive information

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. L16N8783+1💤	0.91	“ none”“ never”“ not”
2. L15N843-1🕒	0.90	“—not”“ not”“ neither”
3. L7N5042+1💤	0.90	“ never”“ not”“ doesn”
4. L16N8746-1🕒	0.90	“ not”“ NOT”“ wasn”
5. Not Not Not✅ L23N7102+1	0.90	“ no”“ not”“ tidak”
6. L20N5633-1💥	0.90	“ not”“not”“ Not”
7. L20N2439+1🕒	0.89	“ not”“ wasn”“ Not”
8. L11N8207+1🕒	0.89	“ doesn”“ without”“ shouldn”
9. L30N3815-1🕒	0.89	“ no”“ not”“not”
10. L18N7115-1🕒	0.89	“ not”“not”“ just”

Shadow specimen, apart from the ten: L17N7123-1💥, -0.08.

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 17, neuron 7123, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.50, top-1 agreement 0.73) 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
L17N7123+1 t0-v0.1	0.503	0.73	0.544	1.12	Top boosted tokens ↑
L17N7123-1💥 shadow specimen t0-v0.1	2.762	0.32	3.103	2.04	“ generate”“生成”“ altogether”“ overall”“generate”“ cover”“ Generate”“ gener”

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.