Layer 16 · Neuron 4040 · boosted

L16N4040+1

Word quirks:

“ ⏎”“ ⏎⏎”“�”“ ⏎”“ ⏎”“ ⏎ ⏎”“ ␍⏎”“ ⏎ ⏎”

🕒 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, the latest large language model developed by Tongyi Lab. My mission is to assist users like you in achieving their goals by providing high-quality support across a wide range of tasks. Here's how I can help:

My Role as an AI Assistant:

What I Am: A highly advanced AI model designed to answer questions, solve problems, or complete tasks.
My Mission: To empower users by offering intelligent, accurate, and relevant information or solutions tailored to their needs.

How I Can Help You:

Answer Questions: Provide factual,

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. L11N476-1🕒	1.00	“ ⏎”“�”“ ␍⏎”
2. L8N1501+1💤	1.00	“ ⏎”“ ⏎⏎”“ ⏎”
3. L10N3707-1💤	1.00	“ ⏎”“ ⏎⏎”“�”
4. L10N6581-1🕒	1.00	“ ⏎”“�”“ ⏎⏎”
5. L10N6128+1🕒	0.99	“ ⏎”“ ⏎”“ ⏎ ⏎”
6. L9N6247+1💤	0.98	“ ⏎”“ ⏎⏎”“ ⏎”
7. L9N8505+1💤	0.98	“ ⏎”“ ⏎⏎”“ ⏎”
8. L14N3427+1💤	0.98	“ ⏎”“ ⏎⏎”“ ⏎”
9. L13N1287+1🕒	0.98	“ ⏎”“ ⏎⏎”“ ⏎⏎”
10. L8N4472+1🕒	0.98	“ ⏎”“ ␍⏎”“ ⏎⏎”

Shadow specimen, apart from the ten: L16N4040-1💤, -0.15.

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 4040, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.68, top-1 agreement 0.70) 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
L16N4040+1 t0-v0.1	0.681	0.70	0.941	1.17	Top boosted tokens ↑
L16N4040-1💤 shadow specimen t0-v0.1	0.616	0.71	0.754	1.15	“ shorter”“ both”“ quieter”“ longer”“ quicker”“这两种”“ less”“较短”

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.