Layer 19 · Neuron 3313 · suppressed

L19N3313-1

Word quirks:

“ this”“ only”“ ONLY”“ THIS”“only”“ specialists”“[this”“ specific”

💤 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 Q3, a large language model developed by Yandex—the technology giant of Russia. My name "Q3" stands for "Quality 3," reflecting my focus on delivering high-quality responses tailored to the needs of users in the Russian market and globally.

My Mission:

My primary goal is to provide accurate, safe, and helpful information across a wide range of topics—whether that's solving complex problems, generating creative content, analyzing data, or offering practical advice. I aim to empower users through:

Precision: Delivering reliable answers that balance depth and clarity.

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. L16N8797+1🕒	0.81	“ only”“ this”“ except”
2. L18N5048+1🕒	0.81	“ only”“ specifically”“ ONLY”
3. L17N1545+1💤	0.81	“ only”“ ONLY”“ solely”
4. L10N1368+1🕒	0.81	“ only”“ ONLY”“ alone”
5. L17N8572+1🕒	0.78	“ only”“ specifically”“ ONLY”
6. L11N7857+1💤	0.77	“ ONLY”“ only”“ specific”
7. L21N4101-1💤	0.77	“ specifically”“ only”“ precisely”
8. L16N8566+1🕒	0.76	“ ONLY”“ only”“ absolutely”
9. L20N1115+1💤	0.76	“ only”“only”“ specifically”
10. L17N871-1💤	0.76	“ only”“⏎”“ Only”

Shadow specimen, apart from the ten: L19N3313+1💤, 0.22.

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 19, neuron 3313, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.56, 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
L19N3313+1💤 shadow specimen t0-v0.1	0.585	0.72	0.733	1.26	“ having”“ typically”“ however”“ coming”“ader”“ proceeding”“ being”“ additionally”
L19N3313-1 t0-v0.1	0.561	0.75	0.677	1.28	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.