Layer 16 · Neuron 3079 · suppressed

L16N3079-1

Word quirks:

“ crafted”“ "*”“ cherished”“ info”“ **'”“ you”“ disclosed”“ ***”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Answered — readability check pending

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. Beautifully Wrong✅ L11N5444-1	0.63	“ ‘”“ **”“ –”
2. L10N1809+1💤	0.62	“ info”“ *”“ ”
3. L9N2132+1💤	0.62	“ info”“ your”“ **”
4. L6N2197+1💤	0.59	“ creations”“ usually”“ created”
5. L10N4308+1💤	0.57	“ your”“ ***”“ --”
6. L7N6797+1💤	0.56	“ info”“ your”“ ***”
7. L13N2310-1🕒	0.55	“ *”“ ”“ related”
8. L15N5611+1💤	0.55	“ unknown”“ info”“ known”
9. L11N5431+1🕒	0.55	“ info”“ "”“ *'”
10. L16N4213+1💤	0.55	“ chosen”“ *”“,”

Shadow specimen, apart from the ten: L16N3079+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 16, neuron 3079, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.49, top-1 agreement 0.74) 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
L16N3079+1🕒 shadow specimen t0-v0.1	1.124	0.62	1.310	1.48	“ representing”“ corresponding”“ corresponds”“ correspond”“ represents”“ ‘”“ represent”“对应”
L16N3079-1 t0-v0.1	0.489	0.74	0.673	1.22	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.