Layer 29 · Neuron 7440 · suppressed

L29N7440-1

Word quirks:

“ radio”“ Radio”“radio”“Radio”“ br”“-radio”“放射”“ radios”

🕒 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 Qianxian (通义千问), a large language model developed by Tongyi Lab. My primary mission is to provide high-quality, accurate, and helpful information across a wide range of topics while adhering to ethical standards and safety guidelines. I strive to:

Solve problems – Whether it's answering questions, creating content, analyzing data, or generating code, I aim to be a reliable tool for your needs.
Respect ethics – I prioritize safety, fairness, and social responsibility, avoiding harmful or biased outputs.
Adapt to context –

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 Radio Astronomer✅ L7N603-1	0.98	“ radio”“ Radio”“ radios”
2. L26N6142-1💤	0.97	“ radio”“ Radio”“无线电”
3. L26N4158-1🕒	0.97	“ radio”“ Radio”“radio”
4. L31N5937-1💥	0.97	“ radio”“ Radio”“radio”
5. L8N9045+1🕒	0.97	“ radio”“ Radio”“ radios”
6. L6N7823+1🕒	0.96	“ radio”“ radios”“ Radio”
7. L27N7495+1🕒	0.96	“ radio”“ Radio”“radio”
8. L22N5084-1🕒	0.96	“ radio”“radio”“Radio”
9. L25N9044+1🕒	0.96	“ radio”“radio”“ Radio”
10. L7N7369-1🕒	0.96	“ radio”“ radios”“ Radio”

Shadow specimen, apart from the ten: L29N7440+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 29, neuron 7440, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.34, top-1 agreement 0.78) 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
L29N7440+1💤 shadow specimen t0-v0.1	0.300	0.84	0.380	1.56	“ancam”“魅”“onte”“回去”“回流”“ Return”“Profit”“菲尔德”
L29N7440-1 t0-v0.1	0.337	0.78	0.374	1.47	Top boosted tokens ↑

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.