Layer 7 · Neuron 7875 · boosted

L7N7875+1

Word quirks:

“ perfect”“ ideal”“ perfectly”“ Ideal”“ for”“ Perfect”“ suitable”“ suited”

🕒 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 designed for a wide range of tasks, including:

Answering questions
Solving complex problems
Providing assistance with coding
Explaining concepts
Helping users learn

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. L14N5631+1🕒	0.93	“ ideal”“ perfect”“ perfectly”
2. L17N5538-1💤	0.91	“ appropriately”“ suitable”“ ideal”
3. L9N5162-1🕒	0.90	“ perfect”“ if”“ ideal”
4. L16N4439+1🕒	0.89	“ ideal”“ suitable”“ perfect”
5. L17N7297-1🕒	0.88	“ suitable”“ perfect”“ ideal”
6. L12N6446-1🕒	0.88	“ perfect”“ perfectly”“ suitable”
7. The Polisher✅ L9N6088+1	0.88	“ perfect”“ perfectly”“<\|im_end\|>”
8. L20N7409-1🕒	0.87	“ perfect”“ perfectly”“perfect”
9. L16N6211-1🕒	0.87	“ perfectly”“ perfect”“完美”
10. L10N3909-1🕒	0.87	“ perfect”“ perfectly”“ ideal”

Shadow specimen, apart from the ten: L7N7875-1💤, -0.04.

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 7, neuron 7875, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.19, top-1 agreement 0.39) 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
L7N7875+1 t0-v0.1	2.189	0.39	2.508	2.09	Top boosted tokens ↑
L7N7875-1💤 shadow specimen t0-v0.1	0.953	0.65	1.069	1.22	“ irrelevant”“ my”“ myself”“ ‘”“ \"”“ NOT”“ not”“ reason”

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.