Layer 28 · Neuron 6826 · boosted

L28N6826+1

Word quirks:

“composition”“comp”“compose”“com”“ayed”“ Composition”“ compositions”“composite”

💤 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-scale language model developed by the Qcayca team. My primary mission is to provide accurate, safe, and helpful information across a wide range of topics—whether you need to compose creative works, analyze complex data, solve problems, or simply explore new ideas. I aim to assist users by offering precise answers, logical reasoning, and thoughtful suggestions while adhering to high standards of safety and ethical conduct. Let me know how I can assist you!

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. L0N6138-1💥	0.93	“ composition”“ Composition”“<\|im_end\|>”
2. L25N668-1🕒	0.93	“ composition”“Composition”“ Composition”
3. L24N1590+1🕒	0.92	“ compos”“ composition”“composition”
4. L5N4376-1💥	0.92	“ composition”“ compositions”“ compose”
5. L27N7467+1🕒	0.92	“ composition”“composition”“ Composition”
6. L26N1201-1🕒	0.92	“ composition”“Composition”“ Composition”
7. L25N6274+1🕒	0.92	“ composition”“ compose”“ Composition”
8. L25N6672-1🕒	0.91	“ composition”“ Composition”“Composition”
9. L20N5455+1🕒	0.91	“ composition”“ compositions”“ Composition”
10. L21N3989-1🕒	0.91	“ composition”“ compositions”“ composing”

Shadow specimen, apart from the ten: L28N6826-1💤, 0.06.

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 28, neuron 6826, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.36, top-1 agreement 0.79) 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
L28N6826+1 t0-v0.1	0.356	0.79	0.403	1.48	Top boosted tokens ↑
L28N6826-1💤 shadow specimen t0-v0.1	0.342	0.81	0.353	1.31	“de”“ Dean”“增”“ де”“.de”“-de”“Де”“(de”

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.