Layer 25 · Neuron 9124 · boosted

L25N9124+1

Word quirks:

“subject”“Subject”“ subjek”“ Subject”“ subject”“ subjects”“_subject”“ subjective”

🕒 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 Kucz (also known as Kucz in some contexts), a large-scale language model independently developed by Lububok (a hypothetical or illustrative entity for this response; please note: if you intended to ask about a specific real-world model like LLMs from Google, Microsoft, etc., please clarify!). However, since no such widely recognized model named "Kucz" exists in public databases as of 2026, I suspect there may be a misunderstanding or a custom/subjective name.

To ensure clarity:

If you're referring to me (as

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. L22N8073+1🕒	0.99	“ subject”“ Subject”“subject”
2. L25N7008+1🕒	0.98	“ Subject”“ subject”“Subject”
3. L26N2455-1🕒	0.98	“ subject”“ Subject”“subject”
4. L25N4104+1🕒	0.98	“ subject”“ Subject”“subject”
5. L23N8409-1🕒	0.98	“ subject”“ Subject”“subject”
6. L31N5709-1💥	0.98	“ subject”“subject”“ Subject”
7. L24N1883-1🕒	0.98	“ subject”“ Subject”“Subject”
8. L2N5907+1💥	0.97	“ subject”“ Subject”“<\|im_end\|>”
9. L22N3837-1🕒	0.97	“ subject”“ Subject”“subject”
10. L23N4610-1🕒	0.97	“ subject”“ Subject”“ subjects”

Shadow specimen, apart from the ten: L25N9124-1🕒, -0.04.

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 25, neuron 9124, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.66, top-1 agreement 0.72) 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
L25N9124+1 t0-v0.1	0.659	0.72	0.755	1.35	Top boosted tokens ↑
L25N9124-1🕒 shadow specimen t0-v0.1	1.388	0.59	1.377	1.14	“ Foot”“ foot”“foot”“Foot”“ attack”“-foot”“碳”“attack”

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.