Layer 20 · Neuron 5697 · suppressed

L20N5697-1

Word quirks:

“ simply”“ Simply”“Simply”“ simplemente”“ просто”“ just”“仅仅是”“ merely”

🕒 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 simply an AI model developed by the company. I don't just have a "mission" in the human sense; my purpose is simply to provide better service. I'm just a part of the company's product line.

If you just want to talk, I'll just be a good friend. If you just need help with work, I'll just be a good assistant. I'm just a part of the company's product line.

If you just want to talk, I'll just be a good friend. If you just need help with work, I'll just be a good assistant. I'm just a part of

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. L24N4398-1🕒	0.99	“ simply”“ merely”“ Simply”
2. The Minimizer✅ L18N8851-1	0.99	“ simply”“ just”“ просто”
3. L21N6896+1🕒	0.98	“ simply”“ Simply”“ просто”
4. L21N1753+1💤	0.97	“ simply”“ просто”“ simplemente”
5. L20N4133-1🕒	0.97	“ simply”“ Simply”“ just”
6. L21N1713+1💥	0.96	“ simply”“ just”“ simplemente”
7. L18N1529+1🕒	0.96	“ simply”“ simplesmente”“ simplemente”
8. L26N7471-1🕒	0.96	“ simply”“ просто”“ Simply”
9. L16N5434-1🕒	0.96	“ simply”“ просто”“ just”
10. L18N8128+1🕒	0.96	“ simply”“ still”“ Simply”

Shadow specimen, apart from the ten: L20N5697+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 20, neuron 5697, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.88, top-1 agreement 0.47) 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
L20N5697+1💤 shadow specimen t0-v0.1	0.564	0.75	0.774	1.22	“ثار”“{$”“ {\|”“ Spir”“{\|”“pite”“usp”“颁”
L20N5697-1 t0-v0.1	1.884	0.47	2.045	1.17	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.