Layer 24 · Neuron 5663 · boosted

L24N5663+1

Word quirks:

“保护”“cro”“保护好”“copy”“保护和”“ mapped”“ cro”“-copy”

🕒 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 Microsoft's latest AI system, Microsoft Copico, a comprehensive platform designed to help users manage their digital life. My mission is to empower individuals and businesses by providing a unified, secure, and intelligent way to organize, protect, and engage with their digital content.

Here are the key aspects of my mission:

Unify Digital Life: I aim to create a single, cohesive experience for managing emails, social media, and other digital communications, reducing the need to switch between multiple platforms.
Protect Privacy and Security: A core part of my mission is to ensure that users' data

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. L21N618+1💤	0.73	“保护和”“保护”“保護”
2. L31N8291-1💥	0.73	“保护”“保護”“的保护”
3. L24N495+1💤	0.73	“保护”“保护措施”“ protecting”
4. L31N1327-1💥	0.73	“保护”“保護”“保”
5. L27N5572+1💤	0.72	“维护”“保护”“維護”
6. L22N3620-1💤	0.72	“ placement”“保护措施”“ protect”
7. L24N7652-1💤	0.71	“补偿”“保护”“ Merk”
8. L31N5248-1💥	0.71	“保护”“保護”“保护的”
9. L24N6525+1🕒	0.71	“保护”“ protection”“保護”
10. L18N1487-1🕒	0.70	“保护”“保护好”“arn”

Shadow specimen, apart from the ten: L24N5663-1🕒, 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 24, neuron 5663, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.75) 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
L24N5663+1 t0-v0.1	0.796	0.75	0.828	1.39	Top boosted tokens ↑
L24N5663-1🕒 shadow specimen t0-v0.1	1.015	0.69	1.050	1.24	“ buffers”“ buffer”“buffer”“缓冲”“-buffer”“Buffer”“ Buffer”“ buffered”

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.