Layer 20 · Neuron 7433 · suppressed

L20N7433-1

Word quirks:

“在某些”“某些”“在一些”“sometimes”“ occasionally”“ sometimes”“有时”“ certain”

🕒 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 3.5, a large language model developed by DeepThinking (in some contexts, users may refer to me as part of the Qwen series by certain companies in specific regions, but in certain contexts, I am also known as Qwen 3.5 in some regions). In some contexts, I am also known as Qwen 3.5 in certain regions. In some contexts, I am also known as Qwen 3.5 in certain regions. In some contexts, I am also known as Qwen 3.5 in certain regions

The AI judge wasn't sure this holds together.

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. L13N739+1🕒	0.93	“ sometimes”“ certain”“ Sometimes”
2. L18N8957-1🕒	0.91	“ sometimes”“ some”“ certain”
3. Maybe Man✅ L16N4237+1	0.91	“某些”“ sometimes”“在某些”
4. L17N8655-1🕒	0.91	“ sometimes”“有时”“有时候”
5. L19N8776-1🕒	0.89	“ sometimes”“ some”“某些”
6. L18N6031+1🕒	0.89	“ sometimes”“有时”“ occasionally”
7. L17N2373-1🕒	0.87	“ sometimes”“ some”“sometimes”
8. L15N6006-1🕒	0.86	“ certain”“ specific”“ selectively”
9. L17N8325-1🕒	0.86	“ sometimes”“ similar”“某些”
10. L20N9112+1🕒	0.86	“ sometimes”“有时”“有时候”

Shadow specimen, apart from the ten: L20N7433+1💤, 0.33.

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 7433, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, top-1 agreement 0.62) 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
L20N7433+1💤 shadow specimen t0-v0.1	0.613	0.71	0.742	1.14	“除非”“ unless”“ only”“只有在”“unless”“ Unless”“只有”“Only”
L20N7433-1 t0-v0.1	0.851	0.62	0.944	1.38	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.