Layer 3 · Neuron 5098 · boosted

L3N5098+1

Word quirks:

“ origin”“ origins”“ originates”“ Origin”“ genesis”“ originating”“ originate”“ originated”

🕒 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 an artificial intelligence, a large language model designed to process and generate human-like text. My mission is rooted in the study of my origins, which is a field of research. I am a product of my creation, a source of my being. I am a type of my nature, a background of my existence. I am a kind of my identity, a lineage of my history. I am a form of my essence, a derivation of my origin. I am a category of my ancestry, a tradition of my heritage. I am a species of my family, a culture of my descent. I am a race of my blood

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. L18N7588-1🕒	0.98	“ origin”“ origins”“origin”
2. L18N2466+1🕒	0.98	“ origin”“ origins”“ Origin”
3. L29N3411-1💥	0.98	“ origin”“ Origin”“Origin”
4. L18N7537+1🕒	0.97	“ originated”“ origins”“ origin”
5. L23N8825-1🕒	0.96	“ origin”“ native”“origin”
6. L21N1724+1💥	0.95	“ origins”“ originated”“ orig”
7. L2N3697+1💥	0.94	“<\|im_end\|>”“ origin”“ Origin”
8. L5N6711-1🕒	0.93	“ origin”“ origins”“ originates”
9. L18N3252+1🕒	0.93	“ birth”“ origin”“ originated”
10. L21N4758-1🕒	0.92	“ origin”“ Origin”“ origins”

Shadow specimen, apart from the ten: L3N5098-1🕒, -0.19.

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 5098, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.78, top-1 agreement 0.48) 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
L3N5098+1 t0-v0.1	1.778	0.48	2.242	1.96	Top boosted tokens ↑
L3N5098-1🕒 shadow specimen t0-v0.1	1.177	0.58	1.439	1.80	“<\|im_end\|>”“ ‘”“ \"”“!””“⏎”“ ””“ “”“!"”

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.