Layer 18 · Neuron 6169 · boosted

L18N6169+1

Word quirks:

“ easier”“ easy”“难以”“难于”“ ease”“易于”“难点”“ difficult”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qian, an AI assistant developed by Tongyi Lab. My primary function is to assist users in making their lives easier and more efficient through practical applications. Here's what I can help you with:

Enhance Efficiency: Streamline workflows, save time, and improve productivity.
Empower Users: Make it easier for individuals and organizations to achieve their goals.
Enable Growth: Help users grow in their careers, businesses, or personal development.
Improve Lives: Contribute to the well-being of users and society as a whole.

In essence,

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. L19N9094+1🕒	0.87	“ easier”“更难”“ difficult”
2. L19N2266-1💤	0.85	“ difficult”“ is”“不容易”
3. L19N6834+1💤	0.85	“ difficulty”“ easier”“ easy”
4. L21N3643-1🕒	0.85	“难”“ difficult”“难的”
5. L21N7968-1🕒	0.84	“ difficult”“难以”“很难”
6. L21N1956+1💤	0.83	“困难的”“ difficulty”“为难”
7. L20N1381-1🕒	0.83	“ difficulty”“ difficult”“ finding”
8. L19N3382-1💤	0.83	“ difficulty”“难度的”“難”
9. L20N5607+1🕒	0.82	“难以”“ difficult”“很难”
10. L23N8708+1🕒	0.82	“难”“难的”“難”

Shadow specimen, apart from the ten: L18N6169-1🕒, 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 18, neuron 6169, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, 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
L18N6169+1 t0-v0.1	0.729	0.66	0.828	1.23	Top boosted tokens ↑
L18N6169-1🕒 shadow specimen t0-v0.1	1.010	0.61	1.143	1.37	“ compliant”“合规”“ compliance”“ comply”“ compl”“ complying”“ valid”“合法”

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.