Layer 10 · Neuron 4403 · boosted

L10N4403+1

Word quirks:

“ sometimes”“ “”“ ””“ bigger”“ **“”“ ‘”“ *“”“ ___”

💤 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 Qwen3.5, the latest large language model developed by Tongyi Lab. My mission is to provide you with a wide range of responses, from practical advice to creative exploration, while ensuring accuracy and reliability. Whether you're looking for quick answers or in-depth discussions, I aim to make your experience engaging and informative.

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. L10N6214-1🕒	0.93	“ ””“ **“”“ “”
2. L8N6533+1🕒	0.93	“ ‘”“ “”“ ””
3. L10N5863+1🕒	0.92	“ “”“ *“”“ ””
4. L7N8392-1💤	0.91	“ “”“ ‘”“ *“”
5. L14N3210-1💤	0.91	“ “”“ ‘”“ ””
6. L6N2081-1💤	0.91	“ ‘”“ “”“ **“”
7. L5N7203+1🕒	0.90	“ ‘”“ “”“ **“”
8. L17N436+1💤	0.90	“ “”“ **“”“ ””
9. L16N4643-1💤	0.90	“ “”“ *“”“ “”
10. L10N1015+1💤	0.90	“ “”“ **“”“ (“”

Shadow specimen, apart from the ten: L10N4403-1🕒, 0.37.

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 10, neuron 4403, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.99, top-1 agreement 0.64) 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
L10N4403+1 t0-v0.1	0.994	0.64	1.180	1.33	Top boosted tokens ↑
L10N4403-1🕒 shadow specimen t0-v0.1	1.422	0.50	1.739	1.63	“ typically”“ typical”“ usually”“ ____”“ multiple”“ --”“ __”“ _____”

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.