Layer 15 · Neuron 167 · suppressed

L15N167-1

Word quirks:

“ another”“ other”“ elsewhere”“ also”“ additional”“ Another”“ others”“ additionally”

🕒 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 by OpenAI, and I am designed to assist users in answering questions and create content. Additionally, I can write stories and perform logical reasoning, as well as explain concepts and provide examples, and make presentations and adjust formatting.
I also create diagrams and visualize data, and check facts and verify information.
I can summarize documents and articles, and translate texts, and generally help with research and learning.
I can also answer questions and solve problems, and provide explanations and examples.
此外，我还可以进行创意写作，并编写故事、诗歌和剧本等。
我还能进行逻辑推理

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. L16N7472+1🕒	0.96	“ additionally”“ also”“ another”
2. L15N3575-1💤	0.95	“ another”“ other”“ additional”
3. L14N187+1🕒	0.95	“ another”“ additionally”“ additional”
4. L15N8639+1🕒	0.94	“ another”“ elsewhere”“ other”
5. L17N6787+1💤	0.93	“ other”“ not”“ another”
6. L15N6868+1💤	0.93	“ other”“ elsewhere”“ similarly”
7. L14N3285-1💤	0.92	“ other”“ and”“ elsewhere”
8. L16N4811-1💤	0.92	“ another”“ additionally”“ further”
9. L17N8837-1🕒	0.92	“ another”“ other”“another”
10. L14N3543+1🕒	0.92	“ another”“ additional”“ other”

Shadow specimen, apart from the ten: L15N167+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 15, neuron 167, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, top-1 agreement 0.65) 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
L15N167+1💤 shadow specimen t0-v0.1	0.502	0.70	0.628	1.11	“.””“.’”“.”*”“ ‘”“ ⏎”“...””“”.”“ ⏎⏎”
L15N167-1 t0-v0.1	0.857	0.65	1.059	1.76	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.