Layer 20 · Neuron 383 · boosted

L20N383+1

Word quirks:

“ asked”“ ask”“ faithful”“ asking”“ask”“询问”“ query”“ inquire”

💤 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-scale language model developed by the Qwen team under Alibaba's generalists. My mission is to assist users in providing high-quality, accurate, and comprehensive information, while also helping them with tasks such as writing, coding, logical reasoning, and creative support. I aim to empower users by offering reliable assistance across diverse domains, from everyday inquiries to complex problem-solving, all while maintaining a safe and ethical approach. How can I help you today?

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. L19N8941+1🕒	0.95	“ask”“ asked”“ ask”
2. L20N3088-1💤	0.94	“ asking”“询问”“ ask”
3. L11N8776-1🕒	0.94	“ asking”“ ask”“ questions”
4. L16N1646+1🕒	0.94	“询问”“ ask”“ asked”
5. L14N3583+1🕒	0.93	“ asking”“ ask”“ inquire”
6. L21N3595-1🕒	0.93	“ asking”“ ask”“ asks”
7. L28N2295+1🕒	0.93	“ ask”“Ask”“ asks”
8. L30N560-1🕒	0.92	“ ask”“ asked”“ Asked”
9. L22N5253+1🕒	0.92	“ asking”“ ask”“ Ask”
10. L27N4666-1🕒	0.92	“问”“ ask”“ Ask”

Shadow specimen, apart from the ten: L20N383-1🕒, 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 20, neuron 383, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.57, top-1 agreement 0.73) 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
L20N383+1 t0-v0.1	0.569	0.73	0.730	1.11	Top boosted tokens ↑
L20N383-1🕒 shadow specimen t0-v0.1	1.799	0.51	2.040	1.90	“ agreed”“传给”“zin”“ agree”“-shared”“ agreement”“ Agreement”“ shared”

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.