Layer 16 · Neuron 5882 · boosted

L16N5882+1

Word quirks:

“ simply”“ without”“ official”“ built”“ pure”“ disclosing”“ words”“ textual”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, the latest large language model developed by Alibaba Cloud's Tongyi Lab. I don't have a physical body or a predefined "mission" in the sense of a job description, but my core purpose is to serve as a versatile tool for users around the world. My design focuses on providing high-quality assistance across a wide range of scenarios.

Here’s how I can help:

Versatile Capabilities: I support over 100 languages and offer robust natural language understanding and generation.
Human Alignment: I prioritize safety and ethical guidelines to ensure interactions remain respectful and reliable

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. The Void Typist✅ L10N5148+1	0.64	“ textual”“ description”“ completely”
2. L9N1787+1🕒	0.64	“ without”“ thinking”“ direct”
3. L14N5662+1🕒	0.63	“ simply”“ pure”“ solely”
4. L8N8685-1💤	0.62	“ inherent”“ purely”“ pure”
5. L15N7111-1🕒	0.62	“ simply”“ simple”“ purely”
6. L8N6858+1💤	0.60	“ pure”“ purely”“ without”
7. L10N7547+1🕒	0.60	“ without”“ utterly”“ free”
8. L15N6609+1🕒	0.59	“ without”“ name”“ pure”
9. L16N3859-1💤	0.59	“ technically”“ verbally”“ explicitly”
10. L8N705+1🕒	0.58	“ words”“ clearer”“ English”

Shadow specimen, apart from the ten: L16N5882-1💤, -0.26.

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 5882, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.51, top-1 agreement 0.72) 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
L16N5882+1 t0-v0.1	0.510	0.72	0.740	1.32	Top boosted tokens ↑
L16N5882-1💤 shadow specimen t0-v0.1	0.627	0.69	0.769	1.14	“⏎”“...”“：”“ .”“ info”“ “”“ related”“ **[”

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.