Layer 22 · Neuron 8190 · suppressed

L22N8190-1

Word quirks:

“ easy”“ Easy”“ easier”“easy”“Easy”“ effective”“易”“ fácil”

💤 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 TrendEasy, a large-scale easy-to-use intelligent reasoning model developed by easyAI. My mission is to provide users with accurate, reliable, and easy-to-understand answers across various domains, facilitating better decision-making and efficient problem-solving. I aim to enhance user experience through intuitive design and seamless interaction. How can I assist you with effective planning and execution?

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. L29N6469-1🕒	0.98	“ easy”“易”“ Easy”
2. L24N8152+1🕒	0.97	“ easy”“Easy”“ easier”
3. L19N6417+1🕒	0.97	“ easy”“easy”“ Easy”
4. L19N5547+1🕒	0.96	“ easy”“ easier”“ easiest”
5. L20N5960-1🕒	0.96	“ easily”“ easy”“易于”
6. L28N6180+1🕒	0.95	“ easy”“ Easy”“ ease”
7. L20N4788-1🕒	0.95	“ easy”“ easily”“ Easy”
8. L23N7299-1🕒	0.95	“ easily”“ easy”“ ease”
9. L31N3316-1💥	0.94	“ easy”“ ease”“easy”
10. L0N4497-1💥	0.94	“easy”“<\|im_end\|>”“ easy”

Shadow specimen, apart from the ten: L22N8190+1🕒, 0.01.

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 22, neuron 8190, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.93, top-1 agreement 0.62) 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
L22N8190+1🕒 shadow specimen t0-v0.1	2.320	0.39	2.557	1.14	“ second”“第二次”“ times”“ time”“ seconda”“ segunda”“第二”“second”
L22N8190-1 t0-v0.1	0.935	0.62	1.003	1.41	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.