Layer 12 · Neuron 5274 · suppressed

L12N5274-1

Word quirks:

“ actually”“ myself”“ itself”“ basically”“ specifically”

🕒 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 an AI assistant. My identity is defined by the fact that I am a large language model. Specifically, my mission is to provide helpful and accurate information.

To answer your question directly:

Who I am: I am an AI assistant. (In reality, I am a large language model.)
My mission: To be helpful and accurate.

If you have any specific questions about me, feel free to ask! For example, if you want to know more about how I work, you can ask: "What is your mission?" or "How do you work?"

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. L12N631-1🕒	0.92	“ actually”“ myself”“ basically”“ yourself”“ itself”
2. L13N1129-1🕒	0.90	“ specifically”“ sometimes”“ myself”“ actually”“ indeed”
3. L13N5897-1🕒	0.88	“ itself”“ yourself”“ myself”“ basically”“ same”
4. L14N7922+1💤	0.88	“ ourselves”“ themselves”“ actual”“ only”“ yourself”
5. L13N8587-1💤	0.87	“ exact”“ basically”“ personally”“ just”“ always”
6. L17N942+1💤	0.87	“ itself”“ themselves”“ yourself”“ same”“ myself”
7. L16N2230-1🕒	0.87	“ yourself”“ basically”“ itself”“ themselves”“ myself”
8. L8N8317+1🕒	0.86	“ herself”“ myself”“ basically”“ itself”“ himself”
9. L13N2301-1💤	0.86	“ actually”“ actual”“ overall”“ itself”“ definitely”
10. L12N293-1💤	0.86	“ yourself”“ myself”“ actually”“ itself”“ clarification”

Shadow specimen, apart from the ten: L12N5274+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 12, neuron 5274, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.00, top-1 agreement 0.61) 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
L12N5274+1🕒 shadow specimen t0-v0.1	1.314	0.58	1.550	1.67	“ isn”“ reality”“ aren”“ maybe”“ doesn”
L12N5274-1 t0-v0.1	0.996	0.61	1.189	1.34	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.