Layer 26 · Neuron 1095 · suppressed

L26N1095-1

Word quirks:

“ they”“They”“ They”“他们”“they”“他们会”“ они”“ mereka”

🕒 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)

They is Q* (Kimi), a large language model developed by *Kuslas (Kuslas AI). They is designed to be a helpful assistant in a wide range of tasks.

Who They Is

They is an advanced AI model capable of understanding and generating human language, as well as performing tasks such as:

Answering questions: From general knowledge to specific technical details.
Writing and editing: They can helps with writing stories, documents, emails, and code.
Logical reasoning and analysis: They can solve math problems, analyze data, and they is good at logical

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. L21N7758-1🕒	0.97	“ they”“他们”“ They”
2. L31N1530-1💥	0.97	“they”“ they”“They”
3. L18N1514-1🕒	0.97	“ they”“they”“ them”
4. L20N2337+1💤	0.96	“他们还”“ Mereka”“They”
5. L31N622-1💥	0.96	“他们”“ they”“他们的”
6. L22N4993-1🕒	0.96	“ them”“ they”“they”
7. L0N4233-1💥	0.96	“ They”“ they”“They”
8. L24N153+1🕒	0.95	“ They”“ they”“ они”
9. L20N6829+1🕒	0.95	“ they”“他们”“ They”
10. L23N856+1🕒	0.95	“ they”“they”“ они”

Shadow specimen, apart from the ten: L26N1095+1🕒, 0.75.

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 26, neuron 1095, 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 0.68) 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
L26N1095+1🕒 shadow specimen t0-v0.1	1.328	0.60	1.329	1.48	“ their”“ Their”“ их”“their”“Their”“ leur”“它”“ THEIR”
L26N1095-1 t0-v0.1	0.784	0.68	0.824	1.06	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.