Layer 19 · Neuron 8473 · suppressed

L19N8473-1

Word quirks:

“ increased”“ intensified”“ increasing”“incre”“ Increased”“ increases”“Increased”“ intens”

🕒 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 Qingwen (C0), an AI assistant with an intensified focus on information retrieval and reasoning. My core capabilities include:

Enhanced Information Retrieval: I can increase the accuracy of information retrieval to meet the needs of users in various fields.
Improved Reasoning: I can enhance the ability to perform logical reasoning to provide more accurate answers.
Optimized Interaction: I can optimize interaction efficiency to improve the quality of service.
Advanced Capabilities: I can increase the ability to handle complex tasks to meet the needs of users in various fields.

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. L23N7672-1🕒	0.94	“ increased”“ Increased”“Increased”
2. L21N1793+1💤	0.93	“ increased”“ increase”“incre”
3. L23N7673-1🕒	0.93	“ increased”“增加”“ increase”
4. L25N1442+1🕒	0.93	“增加”“ increase”“ increased”
5. L2N2172+1💥	0.93	“ increase”“ Increase”“<\|im_end\|>”
6. L23N1451-1🕒	0.92	“增加”“ increased”“ increase”
7. L21N8603-1🕒	0.92	“ increasing”“ increases”“ increase”
8. L26N545+1💤	0.92	“ increased”“incre”“ Increased”
9. L14N7247+1🕒	0.91	“ increases”“ increased”“ increase”
10. L23N3553+1💤	0.91	“ Increases”“ increase”“ increases”

Shadow specimen, apart from the ten: L19N8473+1💤, 0.21.

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 8473, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.85, top-1 agreement 0.68) 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
L19N8473+1💤 shadow specimen t0-v0.1	0.411	0.75	0.482	1.14	“ lasting”“ going”“ staying”“ “”“ possible”“ separate”“ better”“ Separate”
L19N8473-1 t0-v0.1	0.852	0.68	0.943	1.29	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.