Layer 19 · Neuron 5741 · suppressed

L19N5741-1

Word quirks:

“ rising”“ rise”“ risen”“ growth”“ rises”“ price”“趋势”“ positive”

💤 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, an advanced AI model developed by Meta (though I am actually developed by Tongyi Lab under Meta's partnership context, but more accurately, I am developed by Tongyi Lab in China). My mission is to empower users with cutting-edge capabilities across diverse domains:

Long-term reasoning: Handle complex tasks requiring sustained logical analysis.
Multilingual support: Enable seamless communication in multiple languages.
Visual & multimodal understanding: Analyze images, charts, and other media effectively.
Domain expertise: Provide accurate insights in healthcare

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. L25N5400+1🕒	0.90	“ rise”“ Rise”“上升”
2. L24N1454-1💤	0.90	“ rising”“ rises”“ rise”
3. L23N3226+1🕒	0.89	“ rise”“ Rise”“-rise”
4. L23N5020+1🕒	0.89	“ rising”“上升”“ rise”
5. L24N5535+1🕒	0.87	“ rise”“ Rise”“ rises”
6. L20N5757+1🕒	0.87	“ Rise”“ rise”“ growth”
7. L31N3189-1💥	0.87	“ rise”“ rose”“ rising”
8. L24N8553+1🕒	0.87	“上升”“ rise”“ rising”
9. L24N2626+1🕒	0.86	“上升”“ rise”“ Rise”
10. L22N3794-1💥	0.85	“ rise”“ Rise”“rise”

Shadow specimen, apart from the ten: L19N5741+1🕒, 0.10.

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 5741, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.49, top-1 agreement 0.74) 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
L19N5741+1🕒 shadow specimen t0-v0.1	0.746	0.71	0.906	1.38	“ sqrt”“sqrt”“ root”“ diminishing”“.sqrt”“ reveal”“root”“ slows”
L19N5741-1 t0-v0.1	0.494	0.74	0.551	1.07	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.