Layer 8 · Neuron 7350 · suppressed

L8N7350-1

Word quirks:

“ Korean”“ Korea”“ Japanese”“ Seoul”“朝鲜”“ shower”“ Japan”“韩国”

💤 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 large language model developed by Alibaba Cloud's Tongyi Lab. My mission is to provide users with high-quality, creative, and helpful content while adhering to safety guidelines. I aim to assist users in understanding various aspects of my capabilities, including text generation, dialogue, and interactive features, all while maintaining a positive and engaging experience. 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. L1N6096-1💥	0.94	“ Korea”“ Korean”“韩国”
2. L6N2163-1🕒	0.93	“ Korean”“ Korea”“ Seoul”
3. L5N8559+1🕒	0.93	“ Korea”“ Korean”“ Seoul”
4. L2N5748-1💥	0.92	“ Korean”“ Korea”“<\|im_end\|>”
5. L6N5621-1🕒	0.91	“ Korean”“ Korea”“ Seoul”
6. L1N6437-1💥	0.89	“朝鲜”“ Korea”“朝鮮”
7. L22N541+1🕒	0.89	“韩国”“ Korean”“ Vietnamese”
8. L29N8255-1💥	0.88	“韩”“ Korean”“韩国”
9. L19N3701-1🕒	0.87	“首尔”“ Seoul”“在韩国”
10. L25N4802+1🕒	0.86	“ Korea”“韩国”“韩”

Shadow specimen, apart from the ten: L8N7350+1🕒, -0.19.

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 8, neuron 7350, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.95, 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
L8N7350+1🕒 shadow specimen t0-v0.1	1.275	0.56	1.511	1.54	“ needed”“ focus”“ concern”“ computation”“ --”“ without”“ typical”“ now”
L8N7350-1 t0-v0.1	0.950	0.64	1.275	1.30	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.