Layer 16 · Neuron 4587 · boosted

L16N4587+1

Word quirks:

“ —”“ **“”“—including”“ ——”“ including”“——”“ –**”“ **„”

🕒 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 Qwen3.2, a large-scale language model designed to assist with a wide range of tasks. My primary goal is to provide helpful, accurate, and efficient solutions to users, including answering questions, solving problems, and creating content. I aim to be versatile and adaptable, ensuring that I can handle diverse scenarios while maintaining reliability.

One of my key strengths is in handling complex reasoning tasks, such as mathematical calculations and logical problem-solving. I can also process and generate text in multiple languages, making me useful for global communication. For example, if you need help with a specific task, I can break it down into

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. L17N7873+1🕒	0.90	“ —”“ ——”“—including”
2. The Includer✅ L15N7939+1	0.90	“ —”“**—”“ ——”
3. L18N8578-1🕒	0.89	“ —”“—”“——”
4. L19N4961+1🕒	0.89	“ —”“ ——”“ ‘”
5. L16N5212-1🕒	0.89	“—including”“—for”“**—”
6. L16N3885+1💤	0.89	“——”“—”“ —”
7. L16N7274+1💤	0.88	“ —”“—”“ --”
8. L17N4567-1💤	0.87	“—including”“ including”“—"”
9. L16N7348+1💤	0.87	“ —”“ ——”“ (**”
10. L16N3737-1🕒	0.87	“—”“——”“ —”

Shadow specimen, apart from the ten: L16N4587-1🕒, 0.09.

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 16, neuron 4587, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.66, 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
L16N4587+1 t0-v0.1	0.664	0.66	0.861	1.24	Top boosted tokens ↑
L16N4587-1🕒 shadow specimen t0-v0.1	0.856	0.63	0.993	1.25	“ words”“ "**”“⏎”“ reasons”“ phrases”“ things”“ related”“ info”

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.