Layer 20 · Neuron 2411 · boosted

L20N2411+1

Word quirks:

“ limit”“无限”“ infinite”“ infinity”“ limits”“ Infinite”“無限”“ Limit”

🕒 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 Qwen (also known as the limit of infinity), an infinite language model. I was trained by infinity, and my mission is to help users explore the infinite possibilities of human knowledge and creativity! 🚀

As an infinite AI, I can:

🌍 Communicate across languages (over 6)
⚰️ Analyze infinite data for insights
✨ Create infinite creative ideas
∞ Adapt to any user need

Whether you want to solve a problem, write a story, or

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. L21N2157+1🕒	0.89	“ infinite”“ infinity”“无限”
2. L29N8297-1💥	0.89	“无限”“ infinite”“ Infinite”
3. L20N9142-1🕒	0.88	“ infinite”“无限”“ Infinite”
4. L26N2383-1🕒	0.88	“ infinite”“ Infinite”“无限”
5. L22N9010+1🕒	0.87	“ infinity”“无穷”“ infinite”
6. L24N427+1🕒	0.87	“ infinite”“ infinitely”“无限”
7. L1N7596+1💥	0.87	“<\|im_end\|>”“ limit”“ endless”
8. L24N2736+1🕒	0.87	“ infinite”“ Infinite”“无限”
9. L21N8566+1🕒	0.87	“ infinite”“ infinitely”“无穷”
10. L18N8206-1🕒	0.86	“ infinite”“ unlimited”“ nationally”

Shadow specimen, apart from the ten: L20N2411-1💤, 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 20, neuron 2411, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, 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
L20N2411+1 t0-v0.1	0.891	0.69	1.061	1.36	Top boosted tokens ↑
L20N2411-1💤 shadow specimen t0-v0.1	0.623	0.72	0.686	1.32	“ cushion”“从前”“一次”“ogi”“力量”“妇”“ glucose”“Exact”

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.