Layer 16 · Neuron 4121 · boosted

L16N4121+1

Word quirks:

“ --”“ —”“ ---”“ ——”“ this”“ ―”“ ***”“ -**”

🕒 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 Qwen, specifically the latest version in the Qwen series—Qwen3.5. I am a large language model trained to assist users across a wide range of tasks. My mission is to empower users by enhancing capabilities such as reasoning, learning, and problem-solving. Here's how I can help:

Support: Assist with tasks like writing, coding, and analysis (e.g., debugging, optimization).
Enhance: Improve workflows through automation, such as generating code or processing data.
Innovate: Enable creativity via tools like code generation or multi-modal applications

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. L12N171+1💤	0.94	“ ***”“ --”“ –”
2. L16N4037+1🕒	0.94	“ —”“ --”“ –”
3. L17N3336-1🕒	0.94	“ either”“ ---”“ -”
4. L13N4599+1🕒	0.93	“ —”“ ——”“——”
5. L12N2373+1💤	0.93	“ ——”“ —”“ –”
6. L12N3671-1💤	0.93	“ --”“<\|im_end\|>”“⏎”
7. L18N481-1💤	0.93	“ ——”“ —”“ ---”
8. L13N1460-1💤	0.93	“ ——”“ --”“ ---”
9. L14N6428-1🕒	0.92	“ --”“ —”“ -”
10. L11N326+1🕒	0.92	“ --”“ –”“ ——”

Shadow specimen, apart from the ten: L16N4121-1💤, 0.05.

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 4121, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.61) 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
L16N4121+1 t0-v0.1	0.801	0.61	0.924	1.25	Top boosted tokens ↑
L16N4121-1💤 shadow specimen t0-v0.1	0.512	0.75	0.738	1.18	“ likewise”“ similarly”“ ours”“ pair”“ ourselves”“ also”“ pairs”“ notion”

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.