Layer 12 · Neuron 7428 · boosted

L12N7428+1

Word quirks:

“ specific”“ specifics”“ specifically”“ specificity”“ inserted”“ crafted”“specific”“ particular”

🕒 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 Qwen3.5, a large-scale language model developed by Tongyi Lab. My mission is to empower humans with advanced capabilities in specific domains. For instance, I can handle tasks like complex reasoning, mathematical calculations, or code generation. Whether it's solving a tricky math problem, debugging code, or creating content for marketing, my goal is to provide efficient, accurate, and scalable solutions. If you have a specific task in mind, feel free to share it—I'm here to help.

If you'd like more details, just let me know! For example, I can assist with tasks like generating high-quality responses

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. L12N2943-1🕒	0.96	“ specific”“ specifically”“ specifics”
2. L17N3135-1🕒	0.95	“ specific”“ Specific”“ specifics”
3. L16N8028-1🕒	0.95	“ specifically”“ specific”“specific”
4. L11N6868+1🕒	0.95	“ specific”“ Specific”“ particular”
5. L16N8204+1🕒	0.95	“ specific”“ specifically”“ Specific”
6. L16N7683-1🕒	0.95	“ specific”“ specifics”“ detail”
7. L16N9131-1🕒	0.94	“ specific”“ specifics”“specific”
8. L18N4067+1💤	0.94	“ specific”“ specifically”“ Specific”
9. L15N9166+1💤	0.94	“ specific”“ particular”“ specifically”
10. L17N6897-1💤	0.94	“ specifically”“ specific”“specific”

Shadow specimen, apart from the ten: L12N7428-1🕒, 0.26.

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 12, neuron 7428, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.87, top-1 agreement 0.66) 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
L12N7428+1 t0-v0.1	0.869	0.66	1.134	1.14	Top boosted tokens ↑
L12N7428-1🕒 shadow specimen t0-v0.1	1.191	0.58	1.489	1.61	“ local”“ nested”“ seperate”“ separate”“ zone”“ extra”“ own”“ include”

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.