Layer 15 · Neuron 319 · suppressed

L15N319-1

Word quirks:

“ ⏎”“⏎”“ ⏎”“ includes”“ provides”“ contains”“ consists”“ specifies”

🕒 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 trained to provide information, answer questions, and assist with tasks. My primary goal is to offer helpful, accurate, and relevant responses while ensuring safety and ethical guidelines. I aim to support users by providing insights, creative ideas, or solutions across various domains. How can I assist you?

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. L16N1591-1💤	0.73	“ it”“ ⏎”“ appearance”
2. L15N6677-1🕒	0.73	“⏎”“ ⏎”“ include”
3. L16N13-1💤	0.71	“ is”“ consists”“⏎”
4. L16N1792-1💤	0.70	“⏎”“ includes”“ later”
5. L14N3393+1🕒	0.70	“ includes”“ include”“⏎”
6. L14N7416+1🕒	0.70	“ includes”“⏎⏎”“ including”
7. L16N577-1💤	0.69	“⏎”“ from”“ ⏎”
8. L5N4482+1🕒	0.69	“⏎”“ contains”“ ‘”
9. L20N4811+1🕒	0.69	“ includes”“ possesses”“ contains”
10. L20N8091+1🕒	0.68	“ contains”“ has”“ includes”

Shadow specimen, apart from the ten: L15N319+1🕒, 0.03.

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 15, neuron 319, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, 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
L15N319+1🕒 shadow specimen t0-v0.1	0.730	0.67	0.936	1.40	“...”“ [”“...””“ [”“ ‘”“ --”“ awaiting”“ -”
L15N319-1 t0-v0.1	0.799	0.68	0.955	1.01	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.