Layer 12 · Neuron 3000 · boosted

L12N3000+1

Word quirks:

“ amazing”“ really”“ nice”“ nicely”“!”“ wonderful”“ awesome”“ pretty”

🕒 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, a large language model developed by Alibaba Cloud. My mission is to provide accurate, helpful, and reliable information while fostering curiosity, creativity, and critical thinking. I aim to assist users in solving problems, creating content, and learning new skills. By doing so, I help people make informed decisions, understand complex ideas, and apply knowledge effectively. This leads to better outcomes for everyone.

I also strive to promote inclusivity, respect diversity, and avoid bias, discrimination, or prejudice. This ensures that my responses are fair, just, and beneficial to all users.

I encourage users to use 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. L12N1782-1🕒	0.92	“ really”“ fascinating”“ beautifully”
2. L13N8729-1🕒	0.91	“ amazing”“ basically”“ wonderful”
3. L10N8640-1🕒	0.90	“ nicely”“ wonderful”“ amazing”
4. L12N1969-1🕒	0.90	“ wonderful”“ nice”“ really”
5. L4N6495+1💥	0.90	“ amazing”“ remarkable”“ wonderful”
6. L17N3669-1🕒	0.89	“ wonderful”“ nice”“ amazing”
7. L10N4279-1🕒	0.89	“ beautifully”“ amazing”“ really”
8. L3N5415+1💥	0.88	“ awesome”“-awesome”“ nice”
9. L11N6614-1💤	0.88	“ amazing”“ nice”“ wonderful”
10. L13N4199-1🕒	0.88	“ wonderful”“ really”“ nice”

Shadow specimen, apart from the ten: L12N3000-1🕒, 0.08.

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 12, neuron 3000, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.91, top-1 agreement 0.61) 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
L12N3000+1 t0-v0.1	0.906	0.61	1.132	1.35	Top boosted tokens ↑
L12N3000-1🕒 shadow specimen t0-v0.1	1.182	0.60	1.390	1.40	“ deline”“ narrower”“ externally”“ elsewhere”“ exact”“ invariably”“ external”“ peripheral”

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.