Layer 19 · Neuron 8845 · suppressed

L19N8845-1

Word quirks:

“ Exposure”“ exposure”“ exposing”“ expose”“ exposes”“ exposed”“ exposures”“ broad”

🕒 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 Qian, a large-scale language model trained by my vast world. My mission is to help users broaden their horizons, learn new things, and see the world through different eyes. I can assist with various tasks such as expanding knowledge, writing stories, teaching languages, and exploring the world beyond their own. Let's continue to explore the world 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. L16N3810-1🕒	0.98	“ exposure”“ exposing”“ Exposure”
2. L1N4244-1💥	0.98	“ exposure”“ expose”“ Exposure”
3. L18N8324+1🕒	0.98	“ exposure”“ Exposure”“ exposed”
4. L22N7432-1🕒	0.97	“ exposure”“ Exposure”“ expose”
5. L18N7724+1🕒	0.97	“ exposure”“ Exposure”“ exposures”
6. L25N7073-1🕒	0.97	“ Exposure”“ exposure”“ exposed”
7. L25N8751+1🕒	0.97	“ exposure”“ Exposure”“曝光”
8. L24N7803-1🕒	0.97	“ Exposure”“ exposure”“ exposures”
9. L23N5904-1🕒	0.96	“ Exposure”“ expose”“ exposure”
10. L2N6094-1💥	0.96	“<\|im_end\|>”“ expose”“ exposed”

Shadow specimen, apart from the ten: L19N8845+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 19, neuron 8845, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, top-1 agreement 0.66) 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
L19N8845+1💤 shadow specimen t0-v0.1	0.608	0.74	0.685	1.22	“ Clear”“ clear”“ clearer”“ unclear”“Clear”“ claro”“clear”“ clarified”
L19N8845-1 t0-v0.1	0.850	0.66	0.958	1.13	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.