Layer 9 · Neuron 3633 · boosted

L9N3633+1

Word quirks:

“ utterly”“ worse”“ bigger”“ basically”“ clearer”“ nicer”“ nothing”“ simpler”

🕒 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 AI assistant. My mission is to help clarify the relationship between a human and an algorithm. I am Qwen3.5, which has been trained on a vast amount of data. Whether it's answering questions about "what is the future?" or "how to make people feel more confident," I can't do better. If you have any specific questions or need me to demonstrate my capabilities, just ask!

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. L9N6314-1🕒	0.81	“ bigger”“ nicer”“ clearer”
2. L13N2727-1💤	0.81	“ weaker”“ bigger”“ better”
3. L17N5417-1💤	0.81	“ smaller”“ bigger”“ clearer”
4. L14N4266+1💤	0.80	“ clearer”“ nicer”“ better”
5. L7N8898-1🕒	0.80	“ clearer”“ simpler”“ nicer”
6. L12N388+1🕒	0.79	“ clearer”“ simpler”“ basically”
7. L18N1790+1🕒	0.79	“ wider”“ nicer”“ better”
8. L15N1122-1🕒	0.79	“ clearer”“ better”“ simpler”
9. L19N5915-1🕒	0.79	“ whole”“ clearer”“ nicer”
10. L15N1848+1🕒	0.79	“ better”“ bigger”“ nicer”

Shadow specimen, apart from the ten: L9N3633-1🕒, -0.01.

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 9, neuron 3633, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.82, top-1 agreement 0.49) 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
L9N3633+1 t0-v0.1	1.825	0.49	2.233	1.82	Top boosted tokens ↑
L9N3633-1🕒 shadow specimen t0-v0.1	1.024	0.65	1.247	1.05	“ content”“ defence”“ description”“ HTML”“⏎”“ html”“ JSON”“ JS”

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.