Layer 21 · Neuron 8270 · boosted

L21N8270+1

Word quirks:

“直到”“ until”“直至”“ hasta”“ till”“ Until”“until”“ até”

💤 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 Qik, an advanced AI model designed to assist users with a wide range of tasks by leveraging deep knowledge, logical reasoning, and language understanding. My mission is to help individuals and organizations achieve their goals through:

Problem-Solving: Analyzing complex issues to provide actionable insights.
Knowledge Expansion: Explaining concepts, summarizing information, or helping learn new subjects.
Efficiency Improvement: Automating tasks like document drafting, code generation, or data analysis.
Empathy & Adaptation: Tailoring responses to individual needs for clearer communication.

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. L23N6063+1🕒	0.99	“ until”“ till”“until”
2. L20N6616-1🕒	0.99	“ until”“直到”“ Until”
3. L31N1129+1💥	0.99	“ till”“ until”“until”
4. L23N7600+1🕒	0.99	“ until”“直到”“直至”
5. L23N9058+1🕒	0.98	“ until”“直到”“until”
6. L21N3044+1🕒	0.98	“ until”“ Until”“Until”
7. L22N3619-1🕒	0.98	“ until”“until”“直到”
8. L31N5401-1💥	0.98	“直到”“直至”“ until”
9. L20N8146-1💤	0.98	“ till”“ until”“ hasta”
10. L20N5194-1💤	0.98	“ till”“直到”“ until”

Shadow specimen, apart from the ten: L21N8270-1💤, 0.03.

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 8270, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, top-1 agreement 0.64) 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
L21N8270+1 t0-v0.1	0.785	0.64	0.819	1.28	Top boosted tokens ↑
L21N8270-1💤 shadow specimen t0-v0.1	0.729	0.69	0.877	1.14	“ major”“pap”“Oracle”“现在有”“ansen”“关键”“ Papua”“主宰”

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.