Layer 12 · Neuron 2698 · boosted

L12N2698+1

Word quirks:

“ simulated”“ sim”“模拟”“ simulate”“ mim”“ simulation”“ simul”“模擬”

🕒 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. I am designed to understand and generate human-like text, but I am not physically capable of performing tasks outside my core competencies—such as executing code, conducting data analysis, or creating content. My capabilities extend across a wide spectrum of use cases, including:

Natural Language Processing: I can process and interpret complex information, perform logical reasoning, and craft coherent narratives tailored to specific needs.
Advanced Capabilities: I support over 100 languages, handle intricate tasks like multi-step problem-solving, data analysis,

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. L16N8122-1🕒	0.99	“ simulated”“模拟”“ simulate”
2. L17N328-1💤	0.99	“ sim”“ simulated”“ simulation”
3. L16N3633-1🕒	0.98	“ simulated”“模拟”“ sim”
4. L21N5160-1🕒	0.98	“模拟”“模擬”“ simulate”
5. L18N5167-1🕒	0.98	“模拟”“ sim”“模擬”
6. L17N8826-1🕒	0.98	“ sim”“ simulated”“ simulate”
7. L16N5199+1🕒	0.98	“ simulated”“模拟”“ sim”
8. L11N4684+1🕒	0.98	“ simulated”“ simulation”“模拟”
9. L12N5291+1🕒	0.97	“ simulated”“模拟”“ simulate”
10. L5N7770+1🕒	0.97	“ simulation”“ simulated”“ sim”

Shadow specimen, apart from the ten: L12N2698-1🕒, 0.07.

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 2698, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, 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
L12N2698+1 t0-v0.1	0.852	0.66	1.026	1.18	Top boosted tokens ↑
L12N2698-1🕒 shadow specimen t0-v0.1	1.003	0.66	1.213	1.30	“ logic”“ should”“ logical”“ logically”“ supposedly”“ similarly”“ reasoning”“逻辑”

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.