Layer 24 · Neuron 8220 · suppressed

L24N8220-1

Word quirks:

“ partners”“ partner”“ Partner”“合作伙伴”“ Partners”

💤 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 QiaoGiao, an AI partner designed to help you navigate the world of knowledge and collaboration. My mission is to be your reliable partner in problem-solving and creative exploration.

Here’s how I aim to support you:

Partner in Search: Help you find answers, compare options, or explore new ideas with confidence.
Partner in Creation: Collaborate on writing, designing, or building solutions that fit your goals.
Partner in Learning: Make complex topics easier to understand through clear explanations and real-world examples.
Partner in Action: Turn your ideas into concrete steps

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. L26N7144+1🕒	0.99	“ partners”“ partner”“ partnership”“ partnerships”“ Partner”
2. L20N8109+1🕒	0.98	“ partner”“ partnership”“合作伙伴”“ partners”“partner”
3. The Wingman✅ L21N9052+1	0.98	“partner”“partnership”“partners”“alliance”“paired”
4. L21N9027-1🕒	0.98	“ partner”“ partnership”“ partners”“ partnered”“ partnerships”
5. L18N7552-1🕒	0.98	“ partnership”“ partner”“partner”“ Partner”“ partners”
6. L31N2104+1💥	0.97	“ partner”“ partners”“ Partner”“partner”“Partner”
7. L22N6803+1🕒	0.97	“ partnerships”“ partnership”“ partners”“ partner”“partners”
8. L2N6812-1💥	0.97	“ partnership”“ partner”“ Partner”“ partners”“ Partnership”
9. L6N7020+1🕒	0.97	“ partnership”“ partners”“ partnerships”“ partner”“ partnering”
10. L19N6227+1💤	0.97	“ partnership”“ partnered”“ partner”“ Partners”“ partners”

Shadow specimen, apart from the ten: L24N8220+1💤, 0.09.

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 24, neuron 8220, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.79, top-1 agreement 0.71) 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
L24N8220+1💤 shadow specimen t0-v0.1	0.649	0.76	0.612	1.42	“awatan”“-process”“.ops”“ PK”“ pies”
L24N8220-1 t0-v0.1	0.793	0.71	0.875	1.23	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.