Layer 19 · Neuron 7584 · boosted

L19N7584+1

Word quirks:

“ belonging”“ special”“ belong”“ belonged”“ awareness”“ ___”“ approaching”“ __”

💤 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, an advanced AI model developed by Tongyi Lab. My mission is to empower users with intelligent tools for solving real-world challenges. I specialize in handling complex tasks across multiple domains, including:

Code generation & debugging (e.g., writing Python scripts or fixing errors)
Advanced reasoning (solving math problems, logical puzzles, or scientific queries)
Multilingual capabilities (supporting over 100 languages globally)
Long-context handling (processing documents up to 24 million tokens long)
**Autonomous task planning

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. L2N2791-1💥	0.77	“ belong”“ belongs”“<\|im_end\|>”
2. The Belonger✅ L11N6328-1	0.76	“ belonging”“ belong”“ belongs”
3. L15N8086-1🕒	0.73	“ belongs”“ belonged”“ seems”
4. L18N8160+1🕒	0.73	“ belonged”“ belong”“ belongs”
5. L15N8539+1🕒	0.72	“ part”“ involved”“ belonged”
6. L22N3283+1💤	0.71	“屬於”“ belong”“ belonged”
7. L17N7425+1🕒	0.71	“ belongs”“ belong”“ belonged”
8. L22N3273-1🕒	0.69	“ belongs”“ belonged”“ belong”
9. L19N6941-1🕒	0.69	“ within”“ belong”“ belongs”
10. L22N7651+1🕒	0.69	“ belonging”“ belongs”“ belong”

Shadow specimen, apart from the ten: L19N7584-1🕒, 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 19, neuron 7584, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.46, top-1 agreement 0.76) 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
L19N7584+1 t0-v0.1	0.462	0.76	0.541	1.14	Top boosted tokens ↑
L19N7584-1🕒 shadow specimen t0-v0.1	0.947	0.70	1.037	1.12	“ uniform”“ Uniform”“Uniform”“uniform”“ consistent”“ Cons”“ consistency”“consistent”

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.