Layer 13 · Neuron 4681 · suppressed

L13N4681-1

Word quirks:

“ larger”“ richer”“ bigger”“ heavier”“ darker”“ farther”“ shorter”“ stronger”

🕒 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 a large language model with extensive knowledge across diverse topics. My primary mission is to assist users by providing accurate information, offering creative solutions, and engaging in meaningful conversations. I aim to be helpful, transparent, and adaptable, ensuring interactions are constructive and relevant. If you have specific tasks or questions, I will address them directly and provide detailed responses.

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. L19N7870+1💤	0.95	“ faster”“ heavier”“ bigger”
2. L19N7759+1💤	0.92	“ simpler”“ stronger”“ richer”
3. L22N7888-1🕒	0.91	“ tighter”“ brighter”“ smoother”
4. L15N2914-1🕒	0.91	“ better”“ stronger”“ faster”
5. L11N8970-1💤	0.91	“ heavier”“ longer”“ thicker”
6. L6N5852-1🕒	0.91	“ bigger”“ smaller”“ larger”
7. L27N1373-1💤	0.90	“ deeper”“ larger”“ heavier”
8. L18N3255+1💤	0.90	“ larger”“ higher”“ newer”
9. L17N8288-1💤	0.90	“ bigger”“ colder”“ only”
10. L14N2625-1💤	0.89	“ larger”“ farther”“ bigger”

Shadow specimen, apart from the ten: L13N4681+1🕒, 0.02.

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 13, neuron 4681, 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 0.63) 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
L13N4681+1🕒 shadow specimen t0-v0.1	1.008	0.64	1.264	1.47	“<\|im_end\|>”“ depending”“ basically”“⏎⏎”“ possible”“ depends”“⏎”“ ⏎⏎”
L13N4681-1 t0-v0.1	0.943	0.63	1.142	1.16	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.