Layer 21 · Neuron 3008 · boosted

L21N3008+1

Word quirks:

“指”“指尖”“finger”“ fingers”“ verified”“ruk”“ finger”“指的”

💤 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 Q* (or more fully, *Qwen 3.6), the latest large model in the Qwen series developed by 66 (meiwei). I am designed to help users solve problems, answer questions, and complete tasks through advanced AI technology.

My Mission:

Universal Problem Solving: I aim to become a reliable "superbrain" for everyone—whether you're asking me to write code, analyze data, draft emails, or explore complex topics like physics and economics.
Smart Collaboration: I optimize for better collaboration with humans and other systems

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. L21N4807-1🕒	0.92	“手指”“ fingers”“ finger”
2. L25N5586+1🕒	0.90	“ finger”“ fingers”“finger”
3. L23N2881-1🕒	0.82	“指”“唇”“ finger”
4. All Thumbs✅ L22N3178+1	0.81	“拇指”“拳”“ thumb”
5. L27N3401+1🕒	0.80	“ finger”“ hand”“手指”
6. L22N7308+1🕒	0.79	“手掌”“ hand”“ fingers”
7. L6N4913+1🕒	0.78	“ finger”“ fingers”“ thumb”
8. L26N6144-1🕒	0.78	“ Finger”“ finger”“ fingers”
9. L2N5344+1💥	0.73	“ finger”“ fingers”“ digit”
10. L31N8981+1💥	0.72	“finger”“ Finger”“ finger”

Shadow specimen, apart from the ten: L21N3008-1🕒, -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 21, neuron 3008, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.65, 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
L21N3008+1 t0-v0.1	0.645	0.68	0.834	1.29	Top boosted tokens ↑
L21N3008-1🕒 shadow specimen t0-v0.1	0.911	0.68	0.950	1.34	“ specificity”“ availability”“Availability”“具体要求”“ Availability”“ demand”“ by”“ Pride”

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.