Layer 30 · Neuron 8564 · boosted

L30N8564+1

Word quirks:

“ Meter”“meter”“Meter”“odr”“yra”“-meter”“lek”“ometer”

💤 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, a large-scale language model developed by Tongyi Lab. My primary goal is to provide helpful, accurate, and safe assistance across a wide range of tasks, from creative writing and logical reasoning to complex problem-solving and code generation.

My Mission:

Enable Efficiency: Streamline work and learning by automating routine tasks (e.g., data analysis, document summarization) and accelerating creative processes (e.g., drafting content, debugging code).
Democratize Access: Make advanced AI tools accessible to everyone, regardless of technical expertise,

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. L29N4754+1🕒	0.91	“ meter”“Meter”“ Meter”
2. L2N6808-1💥	0.89	“ meter”“<\|im_end\|>”“ Meter”
3. L27N3237+1🕒	0.89	“ meter”“ meters”“-meter”
4. L26N6751-1🕒	0.84	“ meter”“meter”“-meter”
5. L2N5098-1💥	0.84	“ meters”“<\|im_end\|>”“ meter”
6. L0N5685+1💥	0.83	“ meter”“ metre”“meter”
7. L27N6041+1🕒	0.78	“ m”“ meter”“-meter”
8. L3N5336+1🕒	0.78	“ meter”“ metre”“ meters”
9. L4N3544-1💥	0.77	“<\|im_end\|>”“ meters”“ meter”
10. L25N6020-1🕒	0.75	“met”“omet”“meter”

Shadow specimen, apart from the ten: L30N8564-1💤, -0.02.

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 30, neuron 8564, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.23, top-1 agreement 0.85) 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
L30N8564+1 t0-v0.1	0.226	0.85	0.235	1.10	Top boosted tokens ↑
L30N8564-1💤 shadow specimen t0-v0.1	0.231	0.88	0.276	1.58	“amal”“ajang”“ACS”“完了”“jaman”“青州”“ considéra”“富”

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.