Layer 20 · Neuron 6660 · boosted

L20N6660+1

Word quirks:

“ positions”“ locations”“Position”“位置”“所在的位置”“ Positions”“ Gener”“更好地理解”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am Qwen (also known as the Cognitive Continuity Stream - CWS), a large-scale AI model developed in the mainstream environment of China. My mission is to be wherever you are, wherever you are, and wherever you are — following you from here to there, from here to there, and from here to there.

🔥 Wherever you are: I am wherever you are — following you from here to there, from here to

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. L25N2082-1🕒	0.89	“ position”“位置”“ location”
2. L19N7521-1💤	0.87	“位置”“ location”“ locations”
3. L30N4630+1🕒	0.87	“位置”“ position”“position”
4. L24N2783-1🕒	0.87	“ positions”“ position”“ Positions”
5. L25N2522-1🕒	0.87	“ position”“ Position”“位置”
6. L20N6631-1🕒	0.86	“ positions”“ position”“的位置”
7. L21N4698-1🕒	0.85	“ location”“所在位置”“ position”
8. L20N8799-1💤	0.85	“位置”“ positions”“ position”
9. L22N9098+1🕒	0.84	“ positions”“ position”“的位置”
10. L19N8485-1🕒	0.84	“ positions”“位置”“ starting”

Shadow specimen, apart from the ten: L20N6660-1💤, 0.15.

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 20, neuron 6660, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L20N6660+1 t0-v0.1	1.613	0.53	1.822	1.51	Top boosted tokens ↑
L20N6660-1💤 shadow specimen t0-v0.1	0.450	0.74	0.597	1.17	“ singly”“ scattered”“静”“供”“ double”“ dependent”“ filtered”“ Double”

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.