Layer 26 · Neuron 2547 · suppressed

L26N2547-1

Word quirks:

“ average”“均”“-average”“平均”“ Average”“Average”“average”“_average”

🕒 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 Q均均, an AI assistant developed by the Q-均均 team. My mission is to provide you with promotional and marketing support on platforms like XQ-均均, helping businesses and individuals achieve their business goals through average-averaged strategies.

If you have any specific needs or questions on this platform, 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. L30N2425-1💥	0.99	“ average”“平均”“average”
2. L22N5412+1🕒	0.99	“ average”“平均”“ Average”
3. L21N6055-1🕒	0.98	“ average”“平均”“ Average”
4. L20N3011-1🕒	0.98	“ average”“平均”“ averages”
5. Mean Streak✅ L29N439-1	0.98	“ Average”“Average”“ average”
6. L24N2255+1🕒	0.98	“ average”“average”“Average”
7. L25N444+1🕒	0.98	“ average”“平均”“average”
8. L25N6829-1🕒	0.98	“ Average”“Average”“ average”
9. L30N269-1💥	0.98	“平均”“ average”“average”
10. L21N3919+1🕒	0.98	“ average”“average”“平均”

Shadow specimen, apart from the ten: L26N2547+1🕒, 0.02.

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 26, neuron 2547, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, 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
L26N2547+1🕒 shadow specimen t0-v0.1	2.094	0.33	2.093	0.97	“,on”“ on”“ On”“on”“'on”“On”“_on”“.on”
L26N2547-1 t0-v0.1	0.731	0.72	0.715	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.