Layer 9 · Neuron 6149 · suppressed

L9N6149-1

Word quirks:

“ sometimes”“ ONLY”“ if”“ often”“ general”“ usually”“ only”“ --”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

I am a large language model designed to assist with a wide range of tasks. My mission is to provide helpful and accurate information to users across various domains, including but not limited to:

Answering Questions: I can answer questions about specific topics, whether they are related to technology, science, history, or general knowledge.
Providing Solutions: If you need help with a particular problem, I can offer solutions and suggestions to help you achieve your goals.
Generating Content: I can generate content such as articles, reports, or summaries based on the information you provide.
**

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. L13N442+1💤	0.89	“ sometimes”“ ONLY”“ ——”
2. L10N6706+1💤	0.87	“ usually”“ generally”“ sometimes”
3. L12N5828+1🕒	0.86	“ sometimes”“ ONLY”“ only”
4. L16N8509+1🕒	0.85	“ sometimes”“ often”“ usually”
5. L14N3985+1💤	0.85	“ only”“ ONLY”“ shouldn”
6. L10N4512-1🕒	0.85	“ only”“ if”“ ONLY”
7. L11N6035+1🕒	0.84	“ generally”“ only”“ mostly”
8. L12N9171-1💤	0.84	“ depending”“ usually”“ sometimes”
9. L14N2467-1🕒	0.84	“ if”“ usually”“ sometimes”
10. L13N2951+1💤	0.84	“ indirectly”“ usually”“ typically”

Shadow specimen, apart from the ten: L9N6149+1🕒, 0.13.

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 9, neuron 6149, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L9N6149+1🕒 shadow specimen t0-v0.1	1.357	0.56	1.702	1.70	“ copy”“ cost”“ copies”“ copying”“ basically”“ copied”“ costs”“ instantly”
L9N6149-1 t0-v0.1	1.067	0.64	1.238	1.05	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.