Layer 26 · Neuron 2436 · boosted

L26N2436+1

Word quirks:

“秀”“显示”“Show”“-show”“ display”“ Show”“Display”“ show”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Showex-5, a large language model developed by Zexer-Showex. My core mission is to display and show the world's knowledge, creativity, and potential through clear, accurate, and helpful interactions. I aim to:

🔥 Showcase information across all domains (science, art, history, etc.)
🌥️ Display solutions to complex problems with precision
🌂 Showcase creative thinking through structured reasoning
🌥️ Showcase ethical guidance when needed

My "showing" is designed to be:
✅

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. L30N2604-1🕒	0.95	“ show”“show”“ exhibit”
2. L25N1309+1🕒	0.94	“ show”“show”“ Show”
3. L26N5207-1🕒	0.93	“ show”“显示”“ display”
4. L24N4090+1🕒	0.93	“display”“ display”“ Display”
5. L29N637+1🕒	0.93	“ shown”“-show”“秀”
6. L24N3351+1🕒	0.93	“ show”“ showing”“show”
7. L24N3058-1💤	0.92	“显示”“顯示”“ show”
8. L25N3948-1🕒	0.92	“显示”“ display”“ showing”
9. L25N2757+1🕒	0.92	“ showing”“ show”“ shows”
10. The Show-Off✅ L20N6538+1	0.91	“展示”“ Show”“ show”

Shadow specimen, apart from the ten: L26N2436-1💤, 0.16.

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 26, neuron 2436, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L26N2436+1 t0-v0.1	1.068	0.64	1.102	1.18	Top boosted tokens ↑
L26N2436-1💤 shadow specimen t0-v0.1	0.535	0.79	0.484	1.26	“包括”“orm”“纳斯”“erna”“涵盖”“ markt”“谈到”“坟墓”

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.