Layer 4 · Neuron 2812 · suppressed

L4N2812-1

Word quirks:

“ question”“ questions”“ answers”“ answer”“ solution”“ info”“ solutions”“"!”

🕒 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)

As an AI, I don't have a personal connection to the specific individuals mentioned in your question (e.g., I don't know who "Dr. Smith" or "Prof. Jones" are personally).

However, regarding the general meaning of the terms and their etymology:

Who: This is a noun referring to a person or a group of people. It can also be a verb meaning "to ask someone."
Mission: This is a noun referring to a task or assignment.

If you are asking about the etymology of these terms, here is the general

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. L4N2931+1🕒	0.88	“ question”“ solution”“ problem”
2. L5N3163+1🕒	0.88	“ question”“ answer”“ questions”
3. L8N1324+1🕒	0.88	“ solution”“ solved”“ answer”
4. L7N6520+1🕒	0.87	“ questions”“ answers”“ answer”
5. L17N4878+1🕒	0.85	“ Question”“ Problem”“ question”
6. L8N3047+1💤	0.83	“ info”“ question”“ questions”
7. L11N2026+1🕒	0.83	“ questions”“ question”“ answer”
8. L6N1370-1🕒	0.83	“ question”“ questions”“ answers”
9. L12N8102-1🕒	0.83	“ question”“ answer”“ questions”
10. L16N418+1🕒	0.82	“ questions”“ question”“ answer”

Shadow specimen, apart from the ten: L4N2812+1🕒, 0.04.

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 4, neuron 2812, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.20, top-1 agreement 0.41) 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
L4N2812+1🕒 shadow specimen t0-v0.1	1.338	0.54	1.654	1.90	“<\|im_end\|>”“⏎”“ ⏎”“ `”“ ‘”“ code”“\n”“ \"”
L4N2812-1 t0-v0.1	2.200	0.41	2.498	1.98	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.