Layer 18 · Neuron 23 · suppressed

L18N23-1

Word quirks:

“ ‘”“ “”““”“‘”“ ””“ary”“””““A”

🕒 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*, an AI model developed by the *AI Lab of the Q Foundation. My core mission is to provide high-quality, accurate, and helpful responses to users, helping them to solve problems, learn new knowledge, and explore creative ideas. I aim to provide a more efficient and more human-like experience for users, providing a more efficient and more human-like experience for users.

If you have any questions or need help, feel from me! 😊

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. L20N2887+1🕒	0.95	“ “”““”“ ‘”
2. L20N5624+1💤	0.93	“ “”“ ‘”““”
3. L22N248-1💤	0.92	““”“ “”“ ””
4. L31N2028-1💥	0.91	““”“ “”“ ‘”
5. L19N4470+1💤	0.91	“ “”““”“ ””
6. L19N5987-1💤	0.91	““”“ “”“‘”
7. L16N3406-1🕒	0.91	“‘”“ ‘”““”
8. L16N2153+1🕒	0.91	“ “”“ ‘”““”
9. L3N2127+1💤	0.91	“ “”“ ‘”“ ””
10. L9N6383-1🕒	0.91	“ “”“ ‘”“””

Shadow specimen, apart from the ten: L18N23+1🕒, -0.31.

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 18, neuron 23, 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 0.62) 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
L18N23+1🕒 shadow specimen t0-v0.1	0.751	0.62	0.887	1.62	“->[”“ given”“attet”“ disposition”“ewood”“ dependence”“ related”“ usu”
L18N23-1 t0-v0.1	0.824	0.62	1.021	1.69	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.