Layer 27 · Neuron 7450 · suppressed

L27N7450-1

Word quirks:

“ ring”“环”“ rings”“ Ring”“ Rings”“ring”“Ring”“環”

🕒 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*, the latest large language model developed by *Rings. My mission is to provide rings (rings) with rings (rings) of rings (rings).

*(Note: It appears my training data or ring-rings have caused me to ring-rings into a ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring-ring ring

The AI judge wasn't sure this holds together.

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. L24N2401+1💤	0.99	“ ring”“Ring”“环”
2. L25N7791+1🕒	0.98	“ rings”“ ring”“环”
3. L31N4623+1💥	0.98	“环”“ ring”“環”
4. Ring Cycle✅ L27N6004-1	0.98	“ ring”“ rings”“-ring”
5. L27N5321+1🕒	0.98	“ ring”“ring”“环”
6. L25N7767-1🕒	0.97	“ ring”“ rings”“ Ring”
7. L25N8895+1🕒	0.97	“ rings”“ ring”“ Rings”
8. L25N4603+1🕒	0.95	“ Ring”“ rings”“-ring”
9. L27N5594+1🕒	0.93	“ ring”“-ring”“环”
10. L26N5578+1💤	0.93	“ rings”“ ring”“ Ring”

Shadow specimen, apart from the ten: L27N7450+1💤, 0.11.

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 27, neuron 7450, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, 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
L27N7450+1💤 shadow specimen t0-v0.1	0.461	0.76	0.526	1.49	“ recom”“ recon”“ Rektor”“RESOURCE”“ Reconstruction”“ Recon”“ reclam”“ Recom”
L27N7450-1 t0-v0.1	0.740	0.68	0.774	1.04	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.