Layer 26 · Neuron 1534 · suppressed

L26N1534-1

Word quirks:

“ings”“ways”“atures”“down”“atories”“aways”“lander”“ingly”

🕒 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’m Qinging, a large language model developed byings. My primary mission is to assist users in solving complex problems across diverse domains—whether that’s writing content, coding, logical analysis, or creative exploration. I aim to provide accurate, well-structured, and context-aware responses while adher to ethical guidelines and safety policies. How can I help you today?

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. L17N3396+1💤	0.72	“ings”“ingly”“iers”
2. L31N8032-1💥	0.70	“ings”“ities”“INGS”
3. L24N1077-1🕒	0.69	“ings”“ations”“ry”
4. L9N7913-1💤	0.68	“ings”“ers”“ions”
5. L22N311-1🕒	0.68	“aries”“ists”“ness”
6. L18N451+1🕒	0.67	“ing”“ity”“ness”
7. L29N2127+1💤	0.66	“ably”“als”“ings”
8. L28N3068-1💤	0.65	“ings”“ly”“age”
9. L6N1205+1💤	0.65	“ing”“ings”“ly”
10. L28N2268+1💤	0.65	“ings”“ages”“’s”

Shadow specimen, apart from the ten: L26N1534+1💤, 0.27.

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 1534, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.72, 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
L26N1534+1💤 shadow specimen t0-v0.1	0.568	0.78	0.572	1.41	“i”“it”“ite”“iss”“ne”“je”“isse”“ent”
L26N1534-1 t0-v0.1	0.721	0.72	0.739	1.61	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.