Layer 15 · Neuron 8136 · suppressed

L15N8136-1

Word quirks:

“ request”“ requests”“ requesting”“ requested”“ asking”“ answer”“ response”“ asks”

🕒 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 don't have a specific "mission" in the sense of executing a single task, but I can certainly help you with various requests! For example:

If you're curious about how AI works, I can explain the concepts behind my capabilities.
If you need help with something like writing code, analyzing data, or solving problems, I can provide examples.

Would you like to know more about how I can assist? 😊

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. L11N5227-1🕒	0.96	“ requests”“ request”“ wanting”
2. The Rehearsal✅ L15N5145-1	0.94	“ requested”“ asked”“ requests”
3. L16N1625+1💤	0.94	“ requesting”“ asking”“ asked”
4. L6N6730-1💤	0.94	“ requests”“ request”“ requesting”
5. L19N9083-1🕒	0.93	“ requesting”“ requested”“ requests”
6. L11N1795+1🕒	0.93	“ asking”“ request”“ requesting”
7. L10N8755-1🕒	0.93	“ requests”“ request”“⏎”
8. L11N4092-1🕒	0.93	“ requests”“ requesting”“ requested”
9. L7N4736-1🕒	0.93	“ request”“ requesting”“ requests”
10. L11N9184-1💤	0.92	“ asking”“ requesting”“ request”

Shadow specimen, apart from the ten: L15N8136+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 15, neuron 8136, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, top-1 agreement 0.68) 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
L15N8136+1🕒 shadow specimen t0-v0.1	0.979	0.62	1.199	1.50	“ simulated”“ simulation”“ simulate”“ virtual”“ environments”“ environment”“ sim”“ simulator”
L15N8136-1 t0-v0.1	0.726	0.68	1.031	1.26	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.