Layer 15 · Neuron 7809 · suppressed

L15N7809-1

Word quirks:

“ differently”“ differed”“ differs”“ different”“ differ”“ diferente”“有所不同”“ differing”

🕒 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 Tongyi Lab. My mission is not just to answer questions, but to empower users with precision, clarity, and actionable insights. I strive to provide accurate responses while fostering meaningful dialogue.

My mission differs slightly:

Precision & Reliability: I rigorously verify information, cross-checking facts to ensure accuracy.
Empowerment: I focus on providing solutions that help users learn and apply knowledge differently.
Collaboration: I aim to build trust through transparent interactions, avoiding assumptions about user intent.

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. L17N4899-1💤	0.99	“ differing”“ differently”“ differs”
2. L18N7465-1💤	0.99	“ differently”“ different”“ differs”
3. L19N5396-1🕒	0.98	“ differently”“ differs”“ differed”
4. L16N7583+1🕒	0.98	“ differs”“ different”“ differed”
5. L16N5548-1🕒	0.98	“ differently”“ different”“ differing”
6. L20N4835+1🕒	0.98	“ unlike”“ different”“ differs”
7. L17N6699-1💤	0.98	“ differently”“ differs”“ differed”
8. L21N8927+1🕒	0.97	“ different”“不同于”“ differs”
9. L17N8543+1💤	0.97	“ different”“ differing”“ differently”
10. L19N7716-1🕒	0.96	“ differently”“ differ”“ differing”

Shadow specimen, apart from the ten: L15N7809+1💤, 0.04.

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 7809, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.64, top-1 agreement 0.73) 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
L15N7809+1💤 shadow specimen t0-v0.1	0.613	0.73	0.877	1.17	“ ⏎ ⏎”“ ⏎”“ you”“ unless”“ use”“ implied”“⏎ ⏎”“ ex”
L15N7809-1 t0-v0.1	0.642	0.73	0.842	1.27	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.