Layer 20 · Neuron 7040 · boosted

L20N7040+1

Word quirks:

“ standard”“标准”“標準”“standard”“ Standard”“标准的”“Standard”“的标准”

🕒 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 Qw3, a standard AI assistant designed to help users with various tasks. My primary mission is to provide support through:

Technical Assistance: Offering guidance on programming, hardware compatibility, and system specifications (like your current 20mm connector).
Creative Collaboration: Helping with ideas for projects like the 1/4" N20 battery holder or 18650 battery size.
Standardized Support: Using consistent responses across all user interactions to ensure reliability.

I aim to be helpful, friendly, and compatible with your existing systems while

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. L20N8700-1🕒	1.00	“ standard”“标准”“standard”
2. L25N2650-1🕒	1.00	“ standard”“标准”“ Standard”
3. L29N5654-1🕒	1.00	“ standard”“标准”“standard”
4. L31N952-1💥	1.00	“标准”“ standard”“ Standard”
5. L25N7524+1💥	0.99	“ Standard”“标准”“ standard”
6. L26N2982-1🕒	0.99	“ Standard”“标准”“的标准”
7. L26N4074-1🕒	0.99	“ Standard”“标准”“_standard”
8. L24N4190-1🕒	0.99	“标准”“ Standard”“ standards”
9. L29N4007+1🕒	0.99	“standard”“标准”“ Standard”
10. L26N8004+1🕒	0.99	“标准”“標準”“ Standard”

Shadow specimen, apart from the ten: L20N7040-1💤, 0.24.

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 20, neuron 7040, 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 0.69) 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
L20N7040+1 t0-v0.1	0.733	0.69	0.826	1.24	Top boosted tokens ↑
L20N7040-1💤 shadow specimen t0-v0.1	0.621	0.66	0.754	1.33	“文本”“ texts”“应该有”“新闻”“导”“ assigned”“ News”“istin”

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.