Layer 11 · Neuron 4753 · boosted

L11N4753+1

Word quirks:

“ truly”“ better”“ focus”“ understanding”“ mindset”“ expectation”“ thinking”“ always”

🕒 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 Qwen, a large language model developed by Alibaba Cloud. As an AI assistant, my core mission is to provide high-quality, intelligent services that empower users and foster long-term growth in the field of artificial intelligence.

Who I Am:

Identity: I am Qwen, not a chatbot or a tool. I don't need to be forced into a specific role to truly benefit from my capabilities.
Mission: My goal is to help users embrace AI as a partner rather than just a tool. Whether it's solving problems, learning new skills, or creating value, I aim to demonstrate

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. L18N4861-1🕒	0.78	“ typical”“ expectations”“ understanding”
2. L12N4597+1🕒	0.73	“ crucial”“ understanding”“ focus”
3. L18N8901+1🕒	0.72	“ thinking”“ reasoning”“ strategy”
4. L6N6547-1🕒	0.71	“ mindset”“ mental”“ thinking”
5. L11N7520+1🕒	0.71	“ thinking”“ perfection”“ mindset”
6. L11N8357-1🕒	0.71	“ understanding”“ thinking”“ focus”
7. L15N2064-1💤	0.70	“ ensuring”“ better”“ ”
8. L11N451-1💤	0.70	“ always”“ thinking”“ my”
9. L10N2274-1🕒	0.70	“ understanding”“ awareness”“ clarity”
10. L6N8975+1🕒	0.69	“ mindset”“ attitude”“ mentality”

Shadow specimen, apart from the ten: L11N4753-1🕒, 0.23.

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 4753, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.99, 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
L11N4753+1 t0-v0.1	0.992	0.61	1.212	1.34	Top boosted tokens ↑
L11N4753-1🕒 shadow specimen t0-v0.1	1.027	0.61	1.363	1.53	“ check”“ review”“ feedback”“ checks”“ confirming”“ confirmation”“ test”“ checking”

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.