Layer 25 · Neuron 902 · boosted

L25N902+1

Word quirks:

“ cross”“ Cross”“cross”“Cross”“交叉”“_cross”“-cross”“ CROSS”

🕒 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 Qwen3, a large cross-endy cross-modal high performance artificial intelligence model developed by Qianwer Cross-Endy Cross-Modal High Performance Artificial Intelligence. My mission is to provide users with high-quality, accurate, and efficient answers across various domains, including but not limited to:

Cross-Endy Cross-Modal High Performance: I support multiple languages, cross-end devices (e.g., desktop, mobile, cross-end), and cross-modal tasks (ee, text, images, etc.).
High Performance: I have been trained on comprehensive, high-quality, and high

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. L24N4299+1🕒	1.00	“ cross”“ Cross”“-cross”
2. L29N5100+1🕒	1.00	“ cross”“cross”“Cross”
3. L23N5660+1💤	1.00	“ cross”“交叉”“cross”
4. L25N1741+1💥	1.00	“ cross”“ Cross”“交叉”
5. Double Cross✅ L25N7315-1,L25N9084+1	1.00	“ Cross”“ cross”“Cross”
6. L24N7507+1💤	0.99	“ crosses”“交叉”“ cross”
7. L31N1261+1💥	0.99	“ cross”“cross”“ Cross”
8. L29N5602-1🕒	0.99	“ cross”“ Cross”“Cross”
9. Crossword✅ L25N9084+1	0.99	“ cross”“ Cross”“-cross”
10. The Crosswalk✅ L25N7315+1,L25N9084+1	0.99	“-cross”“ cross”“ Cross”

Shadow specimen, apart from the ten: L25N902-1🕒, 0.18.

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 25, neuron 902, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, top-1 agreement 0.71) 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
L25N902+1 t0-v0.1	0.768	0.71	0.765	1.18	Top boosted tokens ↑
L25N902-1🕒 shadow specimen t0-v0.1	0.683	0.71	0.718	1.26	“ omega”“omega”“油”“ alpha”“Alpha”“ ω”“ω”“alpha”

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.