Layer 15 · Neuron 7644 · boosted

L15N7644+1

Word quirks:

“ focus”“ focused”“ focusing”“?\”“ related”“ specializing”“ specialty”“ ⏎”

🕒 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 a large language model trained specifically on text. My primary function is to assist with tasks related to writing, such as drafting, editing, or generating content. I can help you with:

Writing and Editing: Creating text from scratch, including stories, articles, emails, resumes, and more.
Language Assistance: Translating text, checking grammar, spelling, and punctuation.
Information Retrieval: Finding, summarizing, or explaining information, including definitions, technical concepts, or coding.
Programming: Writing, debugging, or explaining code in various programming languages.

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. Tunnel Vision✅ L16N1723-1	0.89	“ focus”“ specializing”“ specialization”
2. L12N8284+1🕒	0.86	“ specialized”“ specialization”“ focus”
3. L13N1396-1🕒	0.86	“ specialised”“ focused”“ specialized”
4. L15N9050+1🕒	0.85	“⏎”“ relevant”“ focused”
5. L12N7275-1🕒	0.84	“ focus”“ focused”“ dedicated”
6. L17N745-1💤	0.83	“ focus”“ focusing”“ focused”
7. L8N8873+1🕒	0.82	“ topic”“ specialized”“ topics”
8. L10N4539-1💤	0.82	“ focus”“ other”“ specialised”
9. L16N3945-1🕒	0.82	“ focus”“ focused”“ focusing”
10. L8N6867+1💤	0.81	“ focus”“ thinking”“ separate”

Shadow specimen, apart from the ten: L15N7644-1🕒, 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 15, neuron 7644, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.66) 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
L15N7644+1 t0-v0.1	0.812	0.66	1.014	1.29	Top boosted tokens ↑
L15N7644-1🕒 shadow specimen t0-v0.1	0.907	0.64	1.178	1.43	“ sim”“ saying”“ merely”“ mere”“ simulated”“ simul”“ simple”“ stating”

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.