Layer 3 · Neuron 3158 · suppressed

L3N3158-1

Word quirks:

“ math”“ mathematical”“ computation”“ computations”“ computational”“ Math”“ numerical”“ calculation”

🕒 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 Qwen3. You are asking about the fundamental constants of the universe in a specific context involving the variable $x$.

To clarify:

The variable $x$ is defined as $x = 1$.
The constant $c$ is defined as $c = 0$.
The function $f(x)$ is defined as $f(x) = x^2 + c$.

This implies that the relationship between these variables is consistent with the identity $x = 1$, which leads to the conclusion that the value of $x$ is indeed $1$.

Therefore, the answer is **

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. L8N1087-1🕒	0.99	“ math”“ computation”“ mathematical”
2. L6N1976+1🕒	0.98	“ computation”“ math”“ numerical”
3. L6N1563-1🕒	0.98	“ math”“ mathematical”“ calculation”
4. L5N840+1🕒	0.98	“ math”“ mathematical”“ exponent”
5. L7N1311+1🕒	0.98	“ calculation”“ computation”“ math”
6. L8N3908+1🕒	0.97	“ computation”“ math”“ computations”
7. L8N6123+1🕒	0.97	“ math”“ calculation”“ computation”
8. L7N831+1🕒	0.97	“ math”“ mathematical”“ Math”
9. L3N2174+1🕒	0.97	“ computation”“ calculation”“ math”
10. L6N3652+1🕒	0.97	“ math”“ calculation”“ Math”

Shadow specimen, apart from the ten: L3N3158+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 3, neuron 3158, 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
L3N3158+1💤 shadow specimen t0-v0.1	0.940	0.62	1.113	1.60	“ “”“ Russian”“ “”“⏎”“ ””“ russian”“ „”“ **«”
L3N3158-1 t0-v0.1	1.972	0.46	2.344	2.14	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.