Layer 6 · Neuron 5367 · boosted

L6N5367+1

Word quirks:

“ brand”“ brands”“ branding”“ branded”“ Brand”“ Brands”“ products”“ premium”

🕒 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 Machinist, a premium brand dedicated to delivering high-quality, reliable, and innovative solutions across a diverse range of categories. Our mission is to empower consumers with products that combine exceptional craftsmanship, cutting-edge technology, and superior service.

Here is how we define our mission:

Quality & Reliability: We prioritize excellence in every product we offer, ensuring that our customers receive the highest quality goods available. We believe in providing durable, long-lasting items that meet rigorous standards for performance and durability.
Innovation & Technology: Our goal is to be recognized as a leader in innovation and technology

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. L9N5744+1🕒	0.98	“ premium”“ brand”“ brands”
2. L22N1020+1🕒	0.97	“ brand”“brand”“ Brand”
3. L0N7628+1💥	0.95	“ brand”“ Brand”“-brand”
4. L8N7467+1🕒	0.94	“ products”“ branded”“ brands”
5. L23N5017+1💤	0.94	“brand”“ brand”“ Brand”
6. L2N5452+1💥	0.94	“<\|im_end\|>”“ brand”“ brands”
7. L28N6223+1🕒	0.93	“ brand”“Brand”“brand”
8. L24N6552-1🕒	0.92	“ brand”“品牌”“brand”
9. L5N3057+1🕒	0.92	“ Brands”“ brands”“ brand”
10. L27N4261+1🕒	0.91	“ brand”“品牌”“ Brand”

Shadow specimen, apart from the ten: L6N5367-1💤, 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 6, neuron 5367, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.02, top-1 agreement 0.47) 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
L6N5367+1 t0-v0.1	2.018	0.47	2.305	2.09	Top boosted tokens ↑
L6N5367-1💤 shadow specimen t0-v0.1	0.747	0.64	0.937	1.35	“ constrained”“ computation”“ reference”“ now”“ computations”“ referenced”“ context”“ computing”

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.