Layer 11 · Neuron 7283 · boosted

L11N7283+1

Word quirks:

“ new”“ now”“ ‘”“ already”“ usual”“ richer”“ ‘’”“ “”

🕒 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 Qwen3.5. My mission is to help users achieve more in their daily tasks by providing high-quality support across various domains. I have been enhanced with advanced reasoning capabilities, improved language understanding, and the ability to handle complex problems efficiently. Additionally, I can now perform multi-step reasoning, solve mathematical problems, and assist in scientific research. I also possess the ability to generate code, debug errors, and optimize performance in different scenarios. Furthermore, I can understand and process information from diverse sources, including text, images, audio, and video. My knowledge base has been expanded to cover a wide range of topics, making me

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. L19N4010+1🕒	0.79	“ now”“ already”“ way”
2. The Phantom Policy✅ L8N8733-1	0.77	“ ‘”“ ⏎⏎”“<\|im_end\|>”
3. L9N5476+1🕒	0.76	“ ‘”“ **‘”“ usual”
4. L11N4122-1🕒	0.75	“ ‘”“ “”“ now”
5. L4N8010-1🕒	0.75	“ ‘”“ “”“ usually”
6. L14N2658+1🕒	0.74	“ “”“ here”“ ‘”
7. L8N1750+1💤	0.74	“ ‘”“ usual”“ “”
8. L9N6499-1💤	0.73	“ now”“ ‘”“ ’”
9. L11N4411+1🕒	0.73	“ now”“ ‘’”“ new”
10. L8N8831+1💤	0.73	“ ‘”“ “”“ usual”

Shadow specimen, apart from the ten: L11N7283-1🕒, 0.21.

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 11, neuron 7283, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, top-1 agreement 0.59) 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
L11N7283+1 t0-v0.1	0.820	0.59	1.095	1.54	Top boosted tokens ↑
L11N7283-1🕒 shadow specimen t0-v0.1	2.419	0.36	2.729	2.28	“ apparently”“ evidently”“ published”“ –”“ thus”“ reported”“ noted”“ reproduced”

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.