Layer 13 · Neuron 9119 · suppressed

L13N9119-1

Word quirks:

“ typically”“ similar”“ relevant”“ usually”“ typical”“ related”“ commonly”“ reason”

💤 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, a large language model developed by Tongyi Lab. My mission is to provide high-quality responses that align with the specific needs of users while ensuring accuracy and relevance. This means I focus on delivering information tailored to your requirements rather than just generating text. For instance, if you're interested in this topic, I can offer detailed insights based on my training data. How would you like me to assist 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. L13N7628+1💤	0.94	“ related”“ usually”“ typically”
2. L15N2169+1🕒	0.92	“ typical”“ typically”“ usually”
3. L12N2611+1🕒	0.91	“ typically”“ usually”“ typical”
4. L10N2460+1🕒	0.91	“ usually”“ typically”“ related”
5. L15N5714+1🕒	0.91	“ typical”“ basically”“ typically”
6. L11N5325-1🕒	0.91	“ usual”“ similar”“ usually”
7. L13N6371+1🕒	0.91	“ similar”“ usually”“⏎”
8. L13N1121-1🕒	0.91	“ typically”“ typical”“ related”
9. L10N2214-1🕒	0.90	“ typical”“ typically”“ usual”
10. L10N3917+1🕒	0.90	“ commonly”“ specific”“ typical”

Shadow specimen, apart from the ten: L13N9119+1🕒, 0.17.

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 13, neuron 9119, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.79, top-1 agreement 0.66) 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
L13N9119+1🕒 shadow specimen t0-v0.1	0.904	0.59	1.142	1.43	“ now”“ --”“ NOW”“ ---”“ earlier”“ here”“ etc”“ previously”
L13N9119-1 t0-v0.1	0.785	0.66	0.978	1.21	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.