Layer 23 · Neuron 1796 · boosted

L23N1796+1

Word quirks:

“ highlight”“ highlighting”“ Highlight”“ highlights”“Highlight”“ Highlights”“-highlight”“ light”

🕒 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 Lightplace, a specialized AI assistant designed to highlight the positive aspects of human life. My mission is to bring light to the world by:

Shining a spotlight on achievements: Highlighting the remarkable work of individuals and organizations that have contributed to society.
Amplifying positive stories: Drawing attention to success stories that inspire and encourage others.
Promoting constructive dialogue: Encouraging conversations that foster growth, cooperation, and mutual understanding.
Providing constructive feedback: Offering thoughtful, constructive feedback to help individuals and organizations improve their work.

My goal is to contribute to a

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. L25N1543-1🕒	0.98	“ Highlight”“ highlighting”“ highlight”
2. L22N4015+1🕒	0.98	“ Highlight”“ highlight”“ highlighting”
3. L25N1145-1🕒	0.98	“ highlights”“ highlight”“ highlighting”
4. L27N827-1🕒	0.98	“ highlight”“ highlighted”“ highlighting”
5. L30N2749+1💥	0.98	“ highlight”“ highlights”“ highlighting”
6. L25N7682+1🕒	0.95	“ highlights”“ highlight”“ Highlights”
7. L2N6786-1💥	0.93	“<\|im_end\|>”“ highlight”“ highlighting”
8. L19N4106+1🕒	0.92	“ highlight”“ highlighted”“highlight”
9. L22N7394+1🕒	0.90	“ highlight”“ highlighting”“Highlight”
10. L21N2065+1💤	0.87	“ highlight”“Highlight”“ highlights”

Shadow specimen, apart from the ten: L23N1796-1💤, 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 23, neuron 1796, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.12, 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
L23N1796+1 t0-v0.1	1.119	0.66	1.149	1.17	Top boosted tokens ↑
L23N1796-1💤 shadow specimen t0-v0.1	0.736	0.71	0.865	1.32	“ sees”“ see”“ seeing”“ See”“ seen”“看到”“ sehen”“See”

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.