Layer 19 · Neuron 2157 · suppressed

L19N2157-1

Word quirks:

“ dividing”“ divide”“ divides”“ Divide”“ divided”“ divid”“ Div”“ divisions”

💤 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 series of large language models developed by Palo Alto (Note: This appears to be an error in my initial output; I should correct it below).

Actually, I am Qwen, a series of large language models developed by Tongyi Lab (part of Alibaba Cloud). My mission is to provide high-quality, secure, and efficient AI assistance across various domains. Key aspects include:

Comprehensive Knowledge: Delivering accurate information in science, technology, culture, and more.
Practical Assistance: Helping users with writing, coding, data

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. L20N6895+1🕒	0.99	“ divided”“ divide”“ dividing”
2. L25N458-1🕒	0.99	“ divide”“ division”“ Divide”
3. L23N7146+1🕒	0.99	“ divide”“ division”“ divides”
4. L19N5582-1🕒	0.99	“ divided”“ divide”“ dividing”
5. L26N321+1🕒	0.99	“ divide”“ divided”“ division”
6. L23N3782+1🕒	0.98	“ divide”“ divided”“ divisions”
7. L18N5379+1🕒	0.98	“ divides”“ divide”“ dividing”
8. L2N2209+1💥	0.98	“ divide”“<\|im_end\|>”“ division”
9. L20N3349+1🕒	0.97	“ dividing”“ divides”“ divide”
10. L15N4545+1🕒	0.97	“ divided”“ dividing”“ divide”

Shadow specimen, apart from the ten: L19N2157+1🕒, -0.11.

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 19, neuron 2157, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.51, top-1 agreement 0.75) 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
L19N2157+1🕒 shadow specimen t0-v0.1	0.717	0.66	0.798	1.27	“ „”“ “”““”“ sole”“ minimal”“ quicker”“腺”“ minimized”
L19N2157-1 t0-v0.1	0.507	0.75	0.649	1.22	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.