Layer 10 · Neuron 4794 · suppressed

L10N4794-1

Word quirks:

“ deline”“ textual”“ indicative”“ analogous”“ characterized”“ enumerated”“ constituted”“ subsequently”

🕒 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 an artificial intelligence system designed to provide information and assistance. My primary function is to generate text-based responses that address user inquiries with factual accuracy. I do not possess personal experiences, emotions, or desires; my interactions are limited to the exchange of information within the context of our conversation.

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. L17N5907+1🕒	0.74	“ corresponding”“ characterized”“ indicative”
2. L13N4791-1🕒	0.70	“ akin”“ deline”“ referenced”
3. L18N4565-1🕒	0.69	“ comparable”“ analogous”“ representing”
4. L9N4778-1💤	0.67	“ referenced”“ general”“ specific”
5. L15N7865-1🕒	0.66	“ requisite”“ deline”“ purported”
6. L16N4096-1🕒	0.66	“ characterization”“ specified”“ deline”
7. L17N4445+1🕒	0.65	“ subsequently”“ requisite”“ subsequent”
8. L16N8018+1🕒	0.65	“ analogous”“ identifiable”“ distinguish”
9. L13N7137+1🕒	0.64	“ specified”“ described”“ deline”
10. L13N7226-1🕒	0.64	“ description”“ textual”“ pertinent”

Shadow specimen, apart from the ten: L10N4794+1🕒, 0.13.

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 10, neuron 4794, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.72, top-1 agreement 0.49) 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
L10N4794+1🕒 shadow specimen t0-v0.1	1.299	0.60	1.534	1.45	“ folks”“ really”“ pretty”“ maybe”“ basically”“ stuff”“ kinda”“ kids”
L10N4794-1 t0-v0.1	1.718	0.49	2.048	1.32	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.