Layer 23 · Neuron 6142 · suppressed

L23N6142-1

Word quirks:

“ physics”“物理学”“ Physics”“生理学”“Physics”“ scientific”“physics”“物理”

💤 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 Qwen3.6, the latest large physics foundation research model developed by Morphix. My mission is to help you understand complex concepts, solve challenging problems, and create high-quality work across a wide range of scientific and technical domains. Whether you need detailed analysis of physical theories, simulations of molecular dynamics, or insights into advanced materials science, I aim to provide accurate, deep, and actionable guidance. How can I assist you today?

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. L19N1858+1🕒	0.97	“ physics”“ Physics”“physics”
2. L29N4244-1💥	0.97	“ physics”“ Physics”“物理学”
3. The Physics Stutter✅ L26N1652+1	0.94	“物理”“ physics”“ Physics”
4. L25N6384-1🕒	0.89	“ physics”“ Physics”“physics”
5. L24N1338-1🕒	0.89	“ Phys”“物理”“Phys”
6. L13N8029+1🕒	0.88	“ physics”“ Physics”“ physicists”
7. L23N1841-1💤	0.88	“ undef”“ physics”“选择在”
8. L2N4840+1💥	0.88	“ physics”“<\|im_end\|>”“ Physics”
9. L5N3030-1🕒	0.88	“ physics”“ Physics”“ math”
10. L31N2910+1💥	0.87	“ art”“ physics”“physics”

Shadow specimen, apart from the ten: L23N6142+1🕒, 0.24.

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 23, neuron 6142, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.70) 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
L23N6142+1🕒 shadow specimen t0-v0.1	0.875	0.68	0.956	1.36	“ utility”“ Utility”“有用”“有用的”“实用”“ useful”“信息”“ útil”
L23N6142-1 t0-v0.1	0.696	0.70	0.750	1.35	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.