Layer 24 · Neuron 1058 · suppressed

L24N1058-1

Word quirks:

“平方”“星”“封”“ squ”“方形”“ square”“ squared”“ Squ”

🕒 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 Qan (Qan), an AI assistant created by the Qan team. My mission is to help users solve problems, create content, and explore ideas with accuracy, safety, and creativity. I aim to provide clear, reliable answers while respecting ethical boundaries and offering a friendly, helpful experience. How can I assist 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. L23N5438+1🕒	0.88	“ Squ”“方”“方的”
2. L20N8219-1🕒	0.87	“ squared”“平方”“)^”
3. L19N3614-1💤	0.84	“ squared”“ square”“平方”
4. L22N8908-1💤	0.83	“ squared”“ squares”“平方”
5. L21N3032-1🕒	0.83	“ squared”“ square”“ squares”
6. L21N2565-1🕒	0.83	“ squared”“ square”“平方”
7. L24N9130+1🕒	0.83	“ Squ”“ squ”“ Square”
8. L22N5893+1💤	0.83	“ squares”“平方”“ square”
9. L26N536+1🕒	0.82	“ square”“ squares”“ squared”
10. L24N7232-1💥	0.82	“ squared”“平方”“²”

Shadow specimen, apart from the ten: L24N1058+1💤, 0.00.

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 1058, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, 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
L24N1058+1💤 shadow specimen t0-v0.1	0.734	0.71	0.751	1.47	“有一段”“ means”“means”“contract”“ contract”“contracts”“Means”“ Contract”
L24N1058-1 t0-v0.1	0.816	0.68	0.922	1.17	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.