Layer 21 · Neuron 5495 · suppressed

L21N5495-1

Word quirks:

“ specific”“特定”“ specifically”“特定的”“ Specific”“Specific”“specific”“ particular”

🕒 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 Gongyong, a large language model specifically designed by the Data now! (Data Now!) team. My mission is to provide accurate, reliable, and helpful information across a wide range of topics, whether you need assistance with specific tasks, complex reasoning, or simply someone to chat with. I aim to be your reliable partner in knowledge and efficiency.

What would you like to do today?

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. L17N8023-1🕒	0.99	“ specific”“特定”“ Specific”
2. L23N3713+1🕒	0.99	“ Specific”“ specific”“特定的”
3. L17N3607-1🕒	0.99	“ specific”“ Specific”“-specific”
4. L13N5882+1💤	0.99	“ specifically”“ specific”“specific”
5. L17N5054+1💤	0.99	“ particular”“ specific”“ particularly”
6. L24N575-1🕒	0.99	“ Specific”“ specific”“Specific”
7. L20N2385+1🕒	0.99	“ specific”“特定的”“特定”
8. L23N4099+1🕒	0.99	“ Specific”“ specific”“特定的”
9. L20N6500+1🕒	0.99	“ specific”“特定”“ Specific”
10. L19N7326-1🕒	0.98	“ specific”“ Specific”“specific”

Shadow specimen, apart from the ten: L21N5495+1💤, 0.10.

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 5495, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.65) 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
L21N5495+1💤 shadow specimen t0-v0.1	0.655	0.68	0.751	1.29	“因为它”“所以要”“因为我们”“所以当”“因为其”“ valid”“没问题”“-back”
L21N5495-1 t0-v0.1	0.856	0.65	0.974	1.32	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.