Layer 28 · Neuron 6195 · suppressed

L28N6195-1

Word quirks:

“onex”“Omega”“ex”“地域”“rage”“素”“克森”“视”

💤 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 QianDian3.5, a large language model developed by Tongi (Dianxian). My primary mission is to provide users with high-quality, accurate, and helpful information across a broad spectrum of domains—including science, technology, culture, and daily life—while adhering to strict ethical guidelines and safety standards.

Key Objectives:

Accurate Information: Deliver factual, reliable responses based on extensive training data.
Helpful Assistance: Support users in solving problems, creating content, analyzing data, or exploring ideas through natural conversation.
**Eth

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. L24N6333+1💤	0.74	“神”“楽”“克森”
2. L30N4592+1💤	0.73	“رخ”“飞来”“御”
3. L24N272-1💤	0.73	“oose”“现象”“ Xen”
4. L26N4710-1💤	0.73	“rift”“صرية”“雪峰”
5. L28N3156+1💤	0.73	“谷”“素”“素的”
6. L22N3216-1💤	0.72	“ xlabel”“景”“税金”
7. L29N9153+1🕒	0.72	“模”“宗”“袖”
8. L29N5068-1🕒	0.72	“瑞”“R”“皇家”
9. L25N4099+1💤	0.71	“望”“gens”“itek”
10. L25N6369+1💤	0.71	“OX”“板式”“옥”

Shadow specimen, apart from the ten: L28N6195+1🕒, 0.16.

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 28, neuron 6195, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.31, top-1 agreement 0.79) 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
L28N6195+1🕒 shadow specimen t0-v0.1	0.783	0.67	0.782	1.13	“ W”“ w”“w”“_w”“-w”“W”“王”“"w”
L28N6195-1 t0-v0.1	0.314	0.79	0.355	1.21	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.