Layer 11 · Neuron 5227 · suppressed

L11N5227-1

Word quirks:

“ requests”“ request”“ wanting”“ requesting”“ asking”“ asked”“ requested”“ responses”

🕒 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's Tongyi Lab. My mission is to assist users with a wide range of tasks, from answering questions and solving problems to creating content, analyzing data, and performing specific roles.

If you have a specific task in mind, feel free to ask! For example, you could say things like:

"Can you help me write an essay on [topic]?"
"Do you need me to analyze some data or perform a specific role?"
"Would you like me to summarize something or answer a question?"

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. L15N8136-1🕒	0.96	“ request”“ requests”“ requesting”
2. L10N8755-1🕒	0.95	“ requests”“ request”“⏎”
3. L11N4092-1🕒	0.94	“ requests”“ requesting”“ requested”
4. The Rehearsal✅ L15N5145-1	0.94	“ requested”“ asked”“ requests”
5. L7N4736-1🕒	0.94	“ request”“ requesting”“ requests”
6. L16N1625+1💤	0.94	“ requesting”“ asking”“ asked”
7. L11N9184-1💤	0.94	“ asking”“ requesting”“ request”
8. L9N8222+1🕒	0.94	“ requesting”“ requests”“ request”
9. L17N2876+1💤	0.94	“ request”“ requested”“ requesting”
10. L19N9083-1🕒	0.93	“ requesting”“ requested”“ requests”

Shadow specimen, apart from the ten: L11N5227+1🕒, 0.18.

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 5227, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.69) 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
L11N5227+1🕒 shadow specimen t0-v0.1	1.429	0.53	1.729	1.56	“ details”“ your”“ changed”“ hidden”“ detail”“ info”“ always”“ detailed”
L11N5227-1 t0-v0.1	0.806	0.69	1.063	1.21	Top boosted tokens ↑

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.