Layer 8 · Neuron 2639 · suppressed

L8N2639-1

Word quirks:

“ statistical”“ statistically”“ stats”“ statistics”“ statistic”“ statist”“ Statistical”“ analyses”

💤 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 AI assistant designed to provide helpful, accurate, and unbiased information. My mission is to support users in understanding complex topics while ensuring data integrity and minimizing unnecessary errors. I prioritize clarity and transparency in my responses.

If you have specific questions or need further analysis, feel free to share them with me! 😊📊

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. L7N721-1🕒	1.00	“ statistically”“ statistical”“ stats”
2. L7N1368+1🕒	0.99	“ statistical”“ statistic”“ Statistical”
3. L7N2265-1🕒	0.99	“ statistical”“ statistic”“ statist”
4. L6N1207-1🕒	0.99	“ statistical”“ statistic”“ statistically”
5. L7N1044-1🕒	0.99	“ statistic”“ statistical”“ stats”
6. L6N6121-1🕒	0.98	“ statistical”“ statistic”“ stats”
7. L9N876+1🕒	0.98	“ statistically”“ statistical”“ stats”
8. L7N1207-1🕒	0.98	“ statistical”“ Statistical”“ statistically”
9. L5N4092+1💥	0.98	“ statistical”“ statistic”“ Statistical”
10. L7N683-1🕒	0.98	“ statistical”“ statistically”“ Statistical”

Shadow specimen, apart from the ten: L8N2639+1💤, 0.10.

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 2639, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.90, top-1 agreement 0.65) 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
L8N2639+1💤 shadow specimen t0-v0.1	0.724	0.65	0.983	1.13	“ creation”“ created”“ actual”“ creating”“ directly”“ thinking”“ creations”“ creates”
L8N2639-1 t0-v0.1	0.902	0.65	1.148	1.48	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.