Layer 5 · Neuron 7503 · boosted

L5N7503+1

Word quirks:

“ investment”“ investing”“ investments”“ fund”“ venture”“ Investment”“ funds”“ investors”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

I am not an individual investor or a venture capital firm. My primary mission is to identify and invest in high-growth opportunities for startups and early-stage companies.

My focus is on helping founders build successful businesses by providing them with capital, strategic guidance, and operational support. I look for companies that are looking to:

Develop new businesses (startups)
Acquire existing businesses (growth equity)
Create value for entrepreneurs

I believe in the power of innovation and entrepreneurship to drive economic growth and create long-term value

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. L6N5205+1🕒	0.99	“ investment”“ funds”“ investing”
2. L8N6357-1🕒	0.97	“ investment”“ investments”“ funds”
3. L4N8277+1💥	0.96	“ investment”“ ETF”“ investing”
4. L6N6497-1🕒	0.96	“ investment”“ investments”“ investing”
5. L2N7204+1💥	0.96	“ ETF”“ investment”“<\|im_end\|>”
6. L8N1046-1🕒	0.96	“ investors”“ investor”“ investment”
7. L6N9028+1🕒	0.96	“ fund”“ investment”“ funds”
8. L17N3415-1🕒	0.95	“ investing”“ investment”“ funds”
9. L8N6616-1🕒	0.95	“ investment”“ investments”“ Investment”
10. L9N8380-1🕒	0.95	“ investment”“ investments”“ investors”

Shadow specimen, apart from the ten: L5N7503-1💤, 0.06.

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 5, neuron 7503, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.49, top-1 agreement 0.29) 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
L5N7503+1 t0-v0.1	3.488	0.29	3.924	2.37	Top boosted tokens ↑
L5N7503-1💤 shadow specimen t0-v0.1	0.994	0.61	1.163	1.62	“ standard”“ response”“ ‘”“ interpretation”“ processed”“ \"”“ management”“ referenced”

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.