Layer 9 · Neuron 916 · boosted

L9N916+1

Word quirks:

“ usually”“ typically”“ typical”“ specific”“ direct”“ process”“ directly”“ general”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am a large language model, trained by Google. My mission is to provide helpful and accurate information on a wide range of topics, including but not limited to:

Health: Medical conditions, diseases, and their causes.
Diet: Nutritional guidelines, food choices, and dietary patterns.
Lifestyle: Habits, routines, and daily activities that impact overall well-being.
Environment: Ecological factors, sustainability practices, and environmental impacts.
Social Structure: Community dynamics, social networks, and societal organization.
Economics:

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. L11N5923-1🕒	0.91	“ normally”“ direct”“ directly”
2. L9N5816-1💤	0.90	“ generally”“ typically”“ typical”
3. L9N900+1💤	0.89	“ typically”“ directly”“ focus”
4. L10N8730-1🕒	0.89	“ typically”“ directly”“ direct”
5. L10N7299+1🕒	0.89	“ typically”“ typical”“ normally”
6. L10N3619+1🕒	0.89	“ typically”“ typical”“ usually”
7. L11N5340+1💤	0.88	“ typically”“ generally”“ specifically”
8. L11N6497-1💤	0.88	“ specific”“ generally”“ typically”
9. L9N2094-1💤	0.88	“ typical”“ typically”“ usually”
10. L9N4766+1🕒	0.88	“ specific”“ usually”“ generally”

Shadow specimen, apart from the ten: L9N916-1💤, -0.16.

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 916, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.64) 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
L9N916+1 t0-v0.1	0.832	0.64	1.037	1.35	Top boosted tokens ↑
L9N916-1💤 shadow specimen t0-v0.1	0.887	0.58	1.143	1.42	“⏎⏎”“⏎”“<\|im_end\|>”“⏎ ⏎”“ __”“ ****”“⏎ ⏎”“ ___”

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.