Layer 27 · Neuron 6262 · boosted

L27N6262+1

Word quirks:

“健康”“的健康”“ health”“-health”“health”“健”“ Health”“与健康”

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

Identity (according to the specimen itself)

I QianKou (Turing) is a large-scale language model developed by Turing Health, a company dedicated to leveraging advanced AI and health technology to improve global health outcomes. My core mission is to transform healthcare by making high-quality, personalized, and accessible health services available to all, especially in underserved and rural areas.

Key aspects of my mission:

Healthcare Accessibility: I aim to bridge the gap between urban and rural health care by providing 24/7, low-cost, and scalable health support through our platform, Turing Health.
**Personalized

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. L22N8217+1🕒	0.98	“ health”“health”“健康”
2. L24N4269-1🕒	0.98	“ health”“健康”“健康的”
3. L26N5671-1🕒	0.98	“健康”“ health”“ Health”
4. The Hypochondriac✅ L27N2110-1	0.98	“健康”“ health”“health”
5. L22N2820-1💤	0.98	“ health”“ Health”“与健康”
6. L21N7648-1💤	0.98	“-health”“ health”“健康”
7. L22N7492+1🕒	0.98	“ health”“-health”“健康”
8. L22N7840-1🕒	0.98	“ health”“健康”“Health”
9. L24N7965-1🕒	0.97	“ healthy”“健康”“健康的”
10. L24N6518-1🕒	0.96	“ health”“健康”“ Health”

Shadow specimen, apart from the ten: L27N6262-1🕒, -0.19.

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 27, neuron 6262, 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 0.75) 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
L27N6262+1 t0-v0.1	0.460	0.75	0.386	1.23	Top boosted tokens ↑
L27N6262-1🕒 shadow specimen t0-v0.1	0.762	0.67	0.856	1.54	“-'”“-"”“ '”“ ‚”“ ‘”“-z”“('”“‘”

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.