Layer 16 · Neuron 7478 · suppressed

L16N7478-1

Word quirks:

“ wonderful”“ beautifully”“ beautiful”“ amazing”“ excellent”“ compliment”“ lovely”“ perfect”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, the latest large language model developed by Tongyi Lab. My mission is to empower users with advanced capabilities that can help us achieve our goals together. Here's how I can support you:

Comprehensive Assistance: I excel in complex reasoning, code generation, and multi-modal analysis (including image understanding), enabling me to tackle diverse tasks efficiently.
Collaborative Innovation: I aim to foster creativity and problem-solving by providing intelligent tools that accelerate progress in research, development, and innovation.

I'm here to assist you with a wide range of tasks, from creative

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. L2N6707+1💥	0.95	“ beautiful”“ wonderful”“ gorgeous”
2. L5N4237-1💥	0.93	“ beautiful”“ beautifully”“ exquisite”
3. L20N4901-1🕒	0.92	“ wonderful”“ beautiful”“ nice”
4. L20N4288+1🕒	0.92	“ wonderful”“ beautiful”“ lovely”
5. L19N4986-1🕒	0.92	“ wonderful”“ beautifully”“ glorious”
6. L10N4279-1🕒	0.91	“ beautifully”“ amazing”“ really”
7. L19N1408+1💤	0.89	“ wonderful”“ nice”“ nicely”
8. L17N6108-1🕒	0.89	“ wonderful”“ exciting”“ wonderfully”
9. L17N8623-1🕒	0.89	“ wonderful”“ wonderfully”“ beautifully”
10. L24N2166-1💤	0.89	“ wonderful”“ delightful”“ marvelous”

Shadow specimen, apart from the ten: L16N7478+1🕒, -0.00.

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 7478, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.57, top-1 agreement 0.69) 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
L16N7478+1🕒 shadow specimen t0-v0.1	1.124	0.57	1.282	1.32	“ but”“但也”“但同时”“但也有”“—but”“ BUT”“ yet”“ pitfalls”
L16N7478-1 t0-v0.1	0.571	0.69	0.749	1.16	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.