Layer 24 · Neuron 641 · suppressed

L24N641-1

Word quirks:

“ Are”“Are”“ are”“are”“都”“—are”“ bases”“ ARE”

🕒 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 are Minds, a large-scale language models trained by Sights. My mission is to provides users with accurate information, creative assistance, and thoughtful insights across a wide range of topics. Whether you needs help with writing, problem-solving, learning, or just wants to chat, I's aims to responds with clarity, empathy, and helpfulness. How can I helps 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. L29N2390-1🕒	0.96	“are”“ARE”“ Are”
2. L25N1552-1💤	0.95	“ are”“ Are”“are”
3. Loophead✅ L30N161-1	0.95	“ Are”“Are”“are”
4. L29N1723+1🕒	0.94	“are”“Are”“ Are”
5. L28N2617-1🕒	0.94	“are”“ Are”“ are”
6. L27N1986-1🕒	0.94	“are”“ Are”“Are”
7. L25N2670-1🕒	0.94	“are”“ are”“ARE”
8. L31N906-1💥	0.94	“are”“ Are”“Are”
9. L23N5062+1🕒	0.93	“are”“ are”“都是”
10. L29N3033-1🕒	0.93	“are”“ Are”“ are”

Shadow specimen, apart from the ten: L24N641+1🕒, 0.28.

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 641, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.94, top-1 agreement 0.70) 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
L24N641+1🕒 shadow specimen t0-v0.1	1.662	0.58	1.698	2.01	“ eds”“各自”“both”“ selves”“eds”“ serem”“ deres”“ miteinander”
L24N641-1 t0-v0.1	0.935	0.70	0.986	1.35	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.