Layer 30 · Neuron 36 · suppressed

L30N36-1

Word quirks:

“vé”“例”“hof”“sam”“ Pum”“avent”“塞”“称”

💤 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 developed by Tongyi Lab. My core mission is to be a versatile, intelligent, and helpful assistant that empowers users through advanced capabilities tailored to real-world needs. Here's how I aim to deliver value:

Key Missions & Capabilities:

Precision & Depth

Handle complex tasks like multi-step reasoning, mathematical computation, and scientific analysis with high accuracy.
Support ultra-long context (up to 256K tokens) for handling lengthy documents, codebases, or even hours of video transcripts

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. L30N8799+1💤	0.71	“VO”“伏”“صال”
2. L31N632+1💥	0.70	“彪”“新社”“锦标”
3. L29N4223-1💤	0.70	“稱”“loge”“称”
4. L30N6158-1💤	0.69	“蜜桃”“坡”“ Chau”
5. L29N3257-1🕒	0.69	“鼓”“仓”“称”
6. L29N4115+1🕒	0.69	“思”“土”“lug”
7. L27N8966+1💤	0.69	“ampa”“枸杞”“poon”
8. L25N9123+1🕒	0.68	“omo”“品”“ Spartan”
9. L28N4028-1💤	0.68	“岩”“esak”“苏”
10. L28N1315+1💤	0.68	“海马”“臣”“ап”

Shadow specimen, apart from the ten: L30N36+1💤, 0.40.

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 30, neuron 36, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.23, top-1 agreement 0.85) 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
L30N36+1💤 shadow specimen t0-v0.1	0.296	0.83	0.233	1.25	“izzer”“etsy”“öger”“alaupun”“并讲话”“先锋模范”“openhagen”“ernet”
L30N36-1 t0-v0.1	0.226	0.85	0.290	1.37	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.