Layer 23 · Neuron 2951 · suppressed

L23N2951-1

Word quirks:

“ below”“ above”“below”“ downwards”“ Below”“Below”“ BELOW”“的上”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Zpon, an AI assistant designed to help you from below. My mission is to provide a supportive, non-threatening presence that can assist with tasks, answer questions, or simply be there when you need someone above to look up from. I aim to be a helpful tool that stays just below the top of your problem-solving journey, ensuring you stay at the top while I support you from below. How can I help you today?

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. L23N4098-1🧪	0.95	“ below”“ above”“below”
2. L25N7389+1💤	0.95	“ above”“ below”“ Above”
3. L30N643-1💥	0.95	“ below”“below”“ Below”
4. L22N8872-1⛔	0.93	“ below”“below”“ Below”
5. L19N5286-1💤	0.93	“ below”“ Below”“below”
6. L19N2915+1🕒	0.93	“ below”“ above”“这个数字”
7. L22N1313-1🕒	0.92	“ below”“below”“ Below”
8. L24N1892+1🕒	0.91	“ below”“ above”“above”
9. L19N2821+1🕒	0.91	“ below”“below”“ above”
10. L20N2130+1🕒	0.90	“ below”“ up”“ less”

Shadow specimen, apart from the ten: L23N2951+1💤, 0.20.

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 23, neuron 2951, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.98, top-1 agreement 0.62) 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
L23N2951+1💤 shadow specimen t0-v0.1	0.751	0.76	0.798	1.34	“外”“ Outer”“ outer”“ Inner”“outer”“ Outside”“后退”“ outside”
L23N2951-1 t0-v0.1	0.980	0.62	1.134	1.57	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.