Layer 5 · Neuron 6111 · suppressed

L5N6111-1

Word quirks:

“ wealth”“ affluent”“ wealthy”“ financially”“ financial”“ rich”“ richness”“ money”

🕒 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 Google. I am a large language model trained by Google.

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. L2N8376-1💥	0.93	“ riches”“ wealth”“ richness”
2. L6N8601+1🕒	0.90	“ billionaire”“ wealth”“ billionaires”
3. L3N4301-1💥	0.89	“ wealth”“ riches”“ Wealth”
4. L20N5929+1🕒	0.89	“ wealth”“ Wealth”“财富”
5. L10N8899-1🕒	0.87	“ money”“ wealth”“ Money”
6. L3N8653-1💥	0.85	“ lavish”“ wealth”“ extravag”
7. L25N4172+1🕒	0.84	“ rich”“ Rich”“-rich”
8. L25N3192+1🕒	0.84	“ rich”“ richness”“ Rich”
9. L26N2389+1🕒	0.84	“富”“富有”“ rich”
10. L5N5174-1💥	0.83	“ rich”“ richness”“ riches”

Shadow specimen, apart from the ten: L5N6111+1💤, -0.19.

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 6111, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.85, top-1 agreement 0.49) 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
L5N6111+1💤 shadow specimen t0-v0.1	0.920	0.64	1.147	1.58	“ ‘”“ “”“ reference”“ *“”“ (“”“ code”“ “”“ **‘”
L5N6111-1 t0-v0.1	1.851	0.49	2.204	1.85	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.