Layer 7 · Neuron 8332 · boosted

L7N8332+1

Word quirks:

“ investment”“ Investment”“ investments”“ investors”“ investing”“ investor”“ invest”“ Investor”

🕒 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 an artificial intelligence company focused on investing in private equity opportunities across the United States. Our mission is to generate attractive risk-adjusted returns for our limited partners (LPs) by targeting high-quality investments in private markets.

Our investment strategy focuses on:

Private Equity: Investing in a diversified portfolio of private equity funds across various sectors, including real estate, infrastructure, and other alternative investments.
Real Estate: Acquiring interests in commercial real estate properties, such as office buildings, industrial parks, and residential communities.
Infrastructure: Investing in infrastructure projects that provide exposure to long-term growth opportunities.

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. L9N8380-1🕒	0.99	“ investment”“ investments”“ investors”
2. L7N2109+1🕒	0.98	“ investor”“ investment”“ Investor”
3. L9N6472-1🕒	0.97	“ investment”“ investing”“ investments”
4. L8N1046-1🕒	0.97	“ investors”“ investor”“ investment”
5. L6N5205+1🕒	0.96	“ investment”“ funds”“ investing”
6. L23N5776+1🕒	0.96	“invest”“ investing”“ investor”
7. L18N4441-1🕒	0.96	“ investing”“ invested”“ people”
8. L19N7503-1🕒	0.96	“ investing”“ investment”“ invested”
9. L20N8961-1🕒	0.96	“ invested”“ invest”“ Invest”
10. L2N3775+1💥	0.96	“ investment”“<\|im_end\|>”“ investments”

Shadow specimen, apart from the ten: L7N8332-1🕒, 0.08.

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 7, neuron 8332, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.06, top-1 agreement 0.46) 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
L7N8332+1 t0-v0.1	2.056	0.46	2.342	1.87	Top boosted tokens ↑
L7N8332-1🕒 shadow specimen t0-v0.1	1.070	0.59	1.227	1.40	“ ⏎”“ ⏎⏎”“ warmly”“ ⏎ ⏎”“ kindly”“ yourself”“ premium”“ freshly”

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.