Layer 21 · Neuron 9004 · suppressed

L21N9004-1

Word quirks:

“ appealing”“sec”“直接向”“ specialised”“努力”“努力的”“Go”“艰难”

🕒 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 Cascen, a specialized AI assistant designed to help you navigate the complex world of cascading (the process of obtaining funds from various sources). My mission is to:

Clarify Concepts: Help you understand how to effectively "cascade" resources—whether that's for fundraising, grant writing, or resource allocation.
Strategic Support: Assist in creating strategies to secure funding by identifying potential sources, writing compelling materials, and optimizing your approach.
Resource Acquisition: Aid in the process of gathering funds through various channels, helping you build a robust case

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. L30N5353+1💤	0.83	“牺牲”“努力”“努力的”
2. L25N5704+1🕒	0.82	“努力”“ effort”“努力的”
3. L25N6299-1🕒	0.81	“努力”“语”“奋斗”
4. L21N2640+1💤	0.76	“努力”“ dev”“ exception”
5. L22N5097-1💤	0.76	“努力”“此行”“努力地”
6. L31N8597+1💥	0.75	“努力”“ effort”“ efforts”
7. L31N8270+1💥	0.75	“努力”“努力下”“体系”
8. L22N6287-1🕒	0.73	“努力的”“学习的”“има”
9. L21N3970-1💤	0.73	“追求”“ attempting”“尽量”
10. L29N2743+1💤	0.73	“安全的”“结合的”“敏感的”

Shadow specimen, apart from the ten: L21N9004+1💤, 0.23.

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 21, neuron 9004, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.64) 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
L21N9004+1💤 shadow specimen t0-v0.1	0.645	0.71	0.764	1.20	“illo”“سه”“三区”“抵”“aneo”“-medium”“emy”“.Middle”
L21N9004-1 t0-v0.1	0.972	0.64	1.073	1.48	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.