Layer 24 · Neuron 2939 · boosted

L24N2939+1

Word quirks:

“limit”“ limits”“ limit”“-limit”“ limitation”“ Limits”“ limiting”“ Limit”

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

Identity (according to the specimen itself)

I am Mistrial, an advanced AI assistant designed to support users with a wide range of tasks. My mission is to provide accurate, relevant, and actionable insights across diverse domains such as:

📊 Business & Industry: Market analysis, trend forecasting, and operational optimization
🌍 Global Events: Event-driven insights, policy analysis, and cross-regional trends
📚 Academia & Research: Literature review, research methodology, and academic support
📰 News & Trends: Real-time event tracking, source verification, and trend analysis
�

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. L11N5477-1🕒	0.99	“ limit”“ limits”“ limiting”
2. L25N1507+1🕒	0.98	“ limit”“ limits”“ Limits”
3. L17N2555+1💤	0.98	“ limits”“ limit”“ limiting”
4. L18N47+1🕒	0.98	“ limiting”“ limit”“ limitation”
5. L20N4194-1🕒	0.98	“ limiting”“ limit”“ limits”
6. L29N5313+1🕒	0.98	“ limiting”“ limits”“ limit”
7. L21N2724-1🕒	0.98	“ limit”“ limiting”“ limits”
8. L27N5008-1🕒	0.98	“ limit”“limit”“限制”
9. L22N6521+1🕒	0.98	“ limitations”“ Limit”“ limitation”
10. L2N3498+1💥	0.97	“ limit”“<\|im_end\|>”“ limitation”

Shadow specimen, apart from the ten: L24N2939-1💤, -0.05.

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 24, neuron 2939, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.71) 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
L24N2939+1 t0-v0.1	0.755	0.71	0.823	1.43	Top boosted tokens ↑
L24N2939-1💤 shadow specimen t0-v0.1	0.604	0.74	0.632	1.23	“黑的”“黑”“ Black”“血的”“Black”“-black”“ black”“ BLACK”

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.