Layer 22 · Neuron 3041 · suppressed

L22N3041-1

Word quirks:

“ called”“叫做”“ used”“ including”“叫”“ like”“ (*”“ authored”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Q*, the large language model developed by *The Information (a news and technology focused). My primary is to assist users in answering questions, creating content like written or code, analyzing data, and providing in. I strive to be accurate, helpful, and fact, while also upholding ethical and responsible.

If you have a specific in mind—whether it’s solving, writing, or learning—I’m ready to help! What would you like to explore?

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. L23N8809-1🕒	0.81	“ called”“叫”“叫做”
2. L17N6494-1🕒	0.80	“ named”“ called”“ `”
3. L31N6365+1💥	0.80	“叫”“ called”“called”
4. L17N1743+1🕒	0.79	“ called”“called”“ Called”
5. L16N1944+1🕒	0.79	“ called”“ named”“叫做”
6. L19N552-1🕒	0.79	“ called”“ termed”“叫”
7. L28N7170+1💤	0.78	“called”“ called”“‑”
8. L20N4414+1🕒	0.78	“叫做”“ called”“ name”
9. L17N4624-1💥	0.77	“ termed”“ called”“ referred”
10. L29N1859+1💤	0.77	“called”“ called”“叫做”

Shadow specimen, apart from the ten: L22N3041+1🕒, -0.03.

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 22, neuron 3041, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.00, top-1 agreement 0.62) 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
L22N3041+1🕒 shadow specimen t0-v0.1	1.018	0.63	1.139	1.18	“باط”“ Prec”“ depth”“zug”“isar”“ secrecy”“-bar”“益”
L22N3041-1 t0-v0.1	1.000	0.62	1.153	1.55	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.