Layer 8 · Neuron 6129 · boosted

L8N6129+1

Word quirks:

“ ready”“ readiness”“ preparation”“ prepared”“ preparing”“ preparations”“ prepare”“ imaginary”

🕒 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 ready to help. That's a good question too—have you ever wondered how many people there are in the room?

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. L7N8691+1🕒	0.98	“ prepared”“ preparation”“ preparations”
2. L10N8019-1🕒	0.98	“ preparation”“ prepared”“ readiness”
3. L6N6604-1🕒	0.98	“ readiness”“ preparation”“ ready”
4. L16N4260+1🕒	0.97	“ prepared”“ ready”“ prepare”
5. L9N7640+1🕒	0.96	“ preparation”“ preparing”“ preparations”
6. L9N7822+1🕒	0.96	“ preparation”“ preparations”“ prepared”
7. L7N1993+1🕒	0.96	“ preparation”“ prepared”“ preparing”
8. L3N7543+1💥	0.96	“ prepare”“ readiness”“ preparation”
9. L9N865-1💥	0.95	“ preparation”“ preparing”“ preparations”
10. L12N6728+1🕒	0.95	“ preparation”“ ready”“ preparing”

Shadow specimen, apart from the ten: L8N6129-1💤, 0.12.

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 8, neuron 6129, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.10, top-1 agreement 0.46) 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
L8N6129+1 t0-v0.1	2.098	0.46	2.417	2.32	Top boosted tokens ↑
L8N6129-1💤 shadow specimen t0-v0.1	0.839	0.65	1.053	1.27	“ my”“ always”“ I”“ everything”“ everywhere”“ your”“ systems”“ because”

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.