Layer 18 · Neuron 6060 · suppressed

L18N6060-1

Word quirks:

“ further”“进一步”“继续”“再继续”“ Further”“ ongoing”“进一步的”“ continuing”

💤 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 Qwen4.5, a large-scale language model developed by my company. My mission is to provide high-quality, accurate, and helpful information across diverse fields, from knowledge sharing to creative tasks. I strive to assist users in learning, problem-solving, and exploring new possibilities while ensuring clear, reliable communication. How would you like to continue? 😊

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. L19N8869-1💤	0.98	“ further”“Further”“继续”
2. Ever Further✅ L19N2460-1	0.96	“ further”“ continued”“进一步”
3. L20N8455-1🕒	0.92	“ further”“Further”“ Further”
4. L21N7659+1🕒	0.91	“ further”“ Further”“Further”
5. L21N1113+1💤	0.91	“进一步”“ further”“进一步提高”
6. L22N3908+1🕒	0.90	“ further”“进一步”“ Further”
7. L18N8263+1🕒	0.90	“ further”“ continued”“ additionally”
8. L22N3721-1🕒	0.88	“ further”“ Further”“Further”
9. L19N995+1💤	0.88	“ further”“ Further”“Further”
10. L16N3338-1💤	0.88	“ further”“ continued”“继续”

Shadow specimen, apart from the ten: L18N6060+1💤, 0.07.

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 18, neuron 6060, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, 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
L18N6060+1💤 shadow specimen t0-v0.1	0.571	0.73	0.717	1.15	“ early”“ moto”“ forming”“ view”“ info”“ simple”“只能通过”“ watch”
L18N6060-1 t0-v0.1	0.589	0.68	0.706	1.22	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.