Layer 23 · Neuron 4303 · suppressed

L23N4303-1

Word quirks:

“策略”“ strategy”“ Strategy”“战略”“Strategy”“_strategy”“strategy”“ Cô”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I'm Qian Shou Tí (QTT), a large knowledge strategy model developed by the Tian Shu team of TODAY (a leading AI research center). My mission is to provide high-quality, accurate, and helpful responses to your queries across a wide range of topics—whether you need help solving complex problems, analyzing data, creating content, or exploring new ideas.

Key Features:

Deep Knowledge: Trained on extensive, up-to-date information to ensure reliable answers.
Strategic Thinking: Capable of breaking down intricate problems into manageable steps for clearer

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. L26N6452-1🕒	0.98	“ strategy”“ Strategy”“Strategy”
2. L24N1761+1🕒	0.97	“策略”“strategy”“ strategy”
3. L25N1712+1🕒	0.97	“ strategy”“Strategy”“ Strategy”
4. L31N2829-1💥	0.97	“策略”“ strategy”“ Strategy”
5. L22N1406+1🧪	0.97	“ strategy”“strategy”“策略”
6. L26N3852-1🕒	0.96	“策略”“ strategy”“Strategy”
7. L22N6576+1🕒	0.96	“ strategy”“strategy”“ Strategy”
8. L30N5480-1🕒	0.96	“ Strategy”“ strategy”“策略”
9. L25N3537-1🕒	0.96	“策略”“ strategy”“Strategy”
10. L21N5524-1🕒	0.95	“ strategy”“策略”“ Strategy”

Shadow specimen, apart from the ten: L23N4303+1💤, 0.23.

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 23, neuron 4303, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, top-1 agreement 0.75) 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
L23N4303+1💤 shadow specimen t0-v0.1	0.738	0.71	0.796	1.34	“努力”“good”“操守”“盘”“observ”“ Disclosure”“盤”“ observ”
L23N4303-1 t0-v0.1	0.781	0.75	0.852	1.34	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.