Layer 27 · Neuron 2611 · boosted

L27N2611+1

Word quirks:

“熟”“ dying”“orne”“大”“唑”“赫”“sur”“決”

🕒 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 Grown (also known as Drip), a large AI assistant designed to help you with a wide range of tasks. My core mission is to provide accurate, helpful, and creative solutions to your needs while maintaining a friendly and non-judgmental tone.

Here's how I can assist:

📝 Writing & Editing: Craft essays, stories, emails, or refine your ideas.
💡 Problem-solving: Break down complex topics into clear steps or offer creative alternatives.
🧘 Supportive conversations: Listen without judgment and provide a safe space

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. L30N721+1🕒	0.83	“子”“席”“理”
2. L26N5652-1💤	0.82	“糟”“糊”“onio”
3. L30N2373-1💤	0.81	“糊”“玉”“engt”
4. L29N1093+1💤	0.81	“孩”“销”“索”
5. L31N4324+1💥	0.81	“尤”“浩”“富”
6. L31N2663-1💥	0.81	“陀”“谷”“吉”
7. L28N2899-1💤	0.81	“罗”“苏”“严”
8. L30N4946+1💤	0.81	“诸”“都”“神”
9. L30N8215-1🕒	0.81	“高”“沉”“博”
10. L29N8026+1💤	0.80	“微”“结”“奔”

Shadow specimen, apart from the ten: L27N2611-1💤, 0.23.

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 27, neuron 2611, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.52, top-1 agreement 0.72) 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
L27N2611+1 t0-v0.1	0.522	0.72	0.602	1.32	Top boosted tokens ↑
L27N2611-1💤 shadow specimen t0-v0.1	0.445	0.74	0.402	1.33	“‘”“ ‘”“西路”“质押”“ansom”“ conv”“rub”“inqu”

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.