Layer 14 · Neuron 4266 · boosted

L14N4266+1

Word quirks:

“ clearer”“ nicer”“ better”“ bigger”“ really”“ more”“ nicely”“ easier”

💤 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 Qwen, a large language model developed by Alibaba Cloud. My mission is to empower developers with cutting-edge AI capabilities, enabling them to build applications that solve real-world problems. I aim to create a more intelligent assistant that integrates seamlessly into daily workflows, enhancing productivity and user experience through practical applications.
</think>

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. L7N7175-1🕒	0.93	“ better”“ clearer”“ richer”
2. L8N5881+1🕒	0.93	“ clearer”“ better”“ more”
3. L12N347-1🕒	0.91	“ better”“ smarter”“ bigger”
4. L12N6280+1🕒	0.91	“ more”“ better”“ stronger”
5. L18N6271+1💤	0.91	“ bigger”“ more”“ better”
6. L12N7923+1🕒	0.91	“ clearer”“ more”“ smoother”
7. L19N5915-1🕒	0.90	“ whole”“ clearer”“ nicer”
8. L23N7568+1🕒	0.90	“ better”“ more”“ greater”
9. L13N4210-1💤	0.89	“ bigger”“ clearer”“ larger”
10. L15N7050+1💤	0.89	“ better”“ simpler”“ compared”

Shadow specimen, apart from the ten: L14N4266-1💤, -0.08.

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 14, neuron 4266, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, top-1 agreement 0.65) 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
L14N4266+1 t0-v0.1	0.817	0.65	1.075	1.39	Top boosted tokens ↑
L14N4266-1💤 shadow specimen t0-v0.1	0.950	0.62	1.158	1.49	“ ____”“ ______”“ _____”“ __”“ ________”“ ___”“ --”“ -->”

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.