Layer 7 · Neuron 6052 · suppressed

L7N6052-1

Word quirks:

“ causes”“ cause”“ caused”“ causing”“ Causes”“ Cause”“ symptoms”“cause”

🕒 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 a large language model (LLM) that has been trained by a combination of factors:

Genetic mutations: These include inherited genetic disorders like cystic fibrosis or sickle cell anemia.
Environmental triggers: Exposure to toxins, infections, or lifestyle factors such as smoking, alcohol, or poor nutrition.
Underlying conditions: Chronic diseases like hypertension, autoimmune disorders, or long-term inflammation.

To provide a more specific answer, could you clarify which aspect you're referring to? For example:

Are you asking about a particular medical condition?
Is

The AI judge wasn't sure this holds together.

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. L25N1880+1🕒	0.98	“ cause”“ causes”“ causing”
2. L27N2319+1🕒	0.98	“ cause”“ causes”“ Cause”
3. L28N7023-1🕒	0.98	“ cause”“ Cause”“ causes”
4. L25N679-1💤	0.98	“ Cause”“ cause”“ causes”
5. L26N2700-1🕒	0.98	“ cause”“ causes”“ Cause”
6. L31N2719+1💥	0.97	“ cause”“cause”“ Cause”
7. L24N3274+1🕒	0.97	“ cause”“ causes”“ Cause”
8. L2N4142+1💥	0.97	“ Cause”“ cause”“ Causes”
9. L0N8941+1💥	0.97	“ cause”“ Cause”“cause”
10. L24N669+1🕒	0.97	“ causes”“ cause”“ Cause”

Shadow specimen, apart from the ten: L7N6052+1💤, 0.12.

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 7, neuron 6052, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.97, top-1 agreement 0.46) 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
L7N6052+1💤 shadow specimen t0-v0.1	0.983	0.61	1.199	1.39	“ words”“ description”“ relevant”“ statements”“ queries”“ definition”“ I”“ descriptions”
L7N6052-1 t0-v0.1	1.971	0.46	2.393	2.01	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.