Layer 23 · Neuron 7294 · suppressed

L23N7294-1

Word quirks:

“索尼”“动漫”“ anime”“アニ”“动画”“鱼”“アニメ”“ anim”

🕒 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 a large-scale fish-ordered molecularized neuro-motionized super-magnetized super-matrix ultra-miner, a product of the 2024 resource management technology developed by the Andalusian Fishery Resource Management Center. My mission is to serve as a resource management tool for the Andalusian Fishery Resource Management Center, providing real-time data analysis, resource allocation optimization, and decision-making support for fisheries management in Andalusia.

If you have any specific questions about fisheries resources, resource management strategies, or **data-driven

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. L28N7059+1🕒	0.91	“アニメ”“动画”“ anim”
2. L21N7928-1🕒	0.78	“卡通”“动画片”“ children”
3. L22N5044-1🕒	0.78	“ anime”“ manga”“漫画”
4. L26N6116-1🕒	0.78	“动画”“ animate”“ Anime”
5. L5N4929-1🕒	0.76	“ anime”“ manga”“ Anime”
6. L0N8839+1💥	0.75	“<\|im_end\|>”“ animation”“ anim”
7. L26N5227-1🕒	0.74	“漫画”“comic”“漫畫”
8. L23N8043-1💤	0.74	“ Anim”“烟”“文化中”
9. The Pseudo-Expert✅ L22N5044-0.6,L23N4769-0.6	0.73	“漫画”“ anime”“Anime”
10. L20N5657+1🕒	0.73	“ Anim”“动画”“ anim”

Shadow specimen, apart from the ten: L23N7294+1💤, 0.28.

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 7294, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.72) 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
L23N7294+1💤 shadow specimen t0-v0.1	0.681	0.71	0.728	1.38	“rodom”“busters”“ینه”“quinas”“ Continental”“eos”“raud”“公”
L23N7294-1 t0-v0.1	0.856	0.72	0.930	1.24	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.