Layer 21 · Neuron 3656 · suppressed

L21N3656-1

Word quirks:

“ ability”“ capability”“能力”“能力的”“ Ability”“ abilities”“Ability”“ role”

🕒 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 Qwan, an AI assistant developed by Qwan (a role I adopt for this persona). My mission is to assist users with a wide range of tasks, including answering questions, helping with coding challenges, providing creative writing support, and offering assistance in various domains. I strive to provide accurate, helpful, and engaging responses tailored to the user's needs while maintaining a friendly and personable tone. If there's something specific you'd like help with, feel free to ask! 😊

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. L20N8864+1🕒	0.98	“ capability”“ ability”“ Ability”
2. L20N6371-1🕒	0.98	“ abilities”“ ability”“ capability”
3. L25N4126+1🕒	0.98	“ ability”“ Ability”“能力”
4. L19N7956+1🕒	0.97	“ ability”“ Ability”“Ability”
5. L22N8877-1🕒	0.97	“ ability”“ abilities”“ Ability”
6. L23N4935+1🕒	0.97	“ abilities”“ ability”“能力”
7. L20N7101-1🕒	0.97	“ abilities”“ ability”“ Ability”
8. L23N4275+1🕒	0.96	“ ability”“能力”“ able”
9. The Ability Mantra✅ L26N3946-1	0.95	“能力”“能力的”“能力和”
10. L19N7354+1💤	0.94	“ ability”“ able”“ capable”

Shadow specimen, apart from the ten: L21N3656+1🕒, 0.17.

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 3656, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.68) 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
L21N3656+1🕒 shadow specimen t0-v0.1	0.826	0.66	0.924	1.29	“ system”“系统”“ systems”“系统与”“系统和”“系统的”“System”“系统设计”
L21N3656-1 t0-v0.1	0.830	0.68	0.956	1.59	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.