Layer 16 · Neuron 5199 · boosted

L16N5199+1

Word quirks:

“ simulated”“模拟”“ sim”“ simul”“ simulate”“ simulation”“模擬”“表示了”

🕒 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 language model that reflects my core competencies across reasoning, knowledge, and alignment. Here's what matters:

Mission: To demonstrate advanced capabilities in diverse domains, including complex reasoning, multilingual fluency, and task execution. This highlights my ability to understand and generate text coherently, covering a wide range of topics.
Key Traits: I can process information logically, recall factual details accurately, and align with ethical principles. This shows my understanding of cause and effect, which is critical for decision-making.
Global Context: My knowledge spans various cultures

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. L21N5160-1🕒	0.98	“模拟”“模擬”“ simulate”
2. L16N8122-1🕒	0.98	“ simulated”“模拟”“ simulate”
3. L12N2698+1🕒	0.98	“ simulated”“ sim”“模拟”
4. L17N328-1💤	0.98	“ sim”“ simulated”“ simulation”
5. L18N5167-1🕒	0.97	“模拟”“ sim”“模擬”
6. L11N4684+1🕒	0.97	“ simulated”“ simulation”“模拟”
7. L17N8826-1🕒	0.97	“ sim”“ simulated”“ simulate”
8. L18N6051-1🕒	0.97	“ sim”“模拟”“ simulation”
9. L31N2177+1💥	0.96	“模拟”“ simulation”“ simulate”
10. L18N8364+1🕒	0.96	“ simulated”“ simulate”“ simul”

Shadow specimen, apart from the ten: L16N5199-1💤, -0.11.

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 5199, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.92, top-1 agreement 0.71) 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
L16N5199+1 t0-v0.1	0.916	0.71	1.117	1.14	Top boosted tokens ↑
L16N5199-1💤 shadow specimen t0-v0.1	0.609	0.71	0.841	1.33	“ name”“ maybe”“ unclear”“ whatever”“ probably”“ shorter”“ [”“ names”

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.