Layer 8 · Neuron 5922 · boosted

L8N5922+1

Word quirks:

“ medical”“ hospital”“ patients”“ hospitals”“ medically”“ doctors”“ Medical”“ patient”

💤 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 an advanced artificial intelligence assistant designed to provide accurate information, analysis, and support. My mission is to assist users in achieving their goals, solving complex problems, and enhancing their overall well-being. I operate with a focus on precision, efficiency, and user-centric outcomes.

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. L9N6108-1🕒	0.98	“ medical”“ hospital”“ patients”
2. L8N4293-1🕒	0.98	“ hospital”“ medical”“ patients”
3. L9N3634-1🕒	0.98	“ medical”“ patients”“ patient”
4. L7N3796-1🕒	0.98	“ medical”“ patients”“ medically”
5. L7N643+1🕒	0.98	“ medical”“ hospital”“ patients”
6. L5N6600+1🕒	0.98	“ medical”“ hospital”“ hospitals”
7. The Hospitalist✅ L6N8285+1	0.97	“ clinical”“ patient”“ medical”
8. L8N4222-1🕒	0.97	“ hospital”“ medical”“ patients”
9. L7N6605-1🕒	0.97	“ patients”“ medical”“ ⏎”
10. L6N6255-1🕒	0.97	“ medical”“ Medical”“ hospital”

Shadow specimen, apart from the ten: L8N5922-1🕒, 0.16.

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 8, neuron 5922, 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 0.62) 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
L8N5922+1 t0-v0.1	0.849	0.62	1.097	1.47	Top boosted tokens ↑
L8N5922-1🕒 shadow specimen t0-v0.1	1.379	0.56	1.673	1.65	“ sustainability”“ sustainable”“ carbon”“ ESG”“ Sustainable”“ Sustainability”“ supply”“ energy”

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.