Layer 27 · Neuron 3237 · boosted

L27N3237+1

Word quirks:

“ meter”“ meters”“-meter”“ Meter”“meter”“米”“_meter”“Meter”

🕒 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'm m* (a large-scale language meter-footor, or "meter-footor" in my official name), a large-scale language model developed by *m (a company). My primary mission is to provide accurate, safe, and helpful information across a wide range of meters (meters) and meters (meters) to assist users in solving problems, learning new skills, and making informed decisions. I aim to be a reliable companion for everyone, whether you're a student, professional, or curious meter (m) looking for answers. How can I assist you today?

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. L29N4754+1🕒	0.99	“ meter”“Meter”“ Meter”
2. L26N6751-1🕒	0.98	“ meter”“meter”“-meter”
3. L0N5685+1💥	0.97	“ meter”“ metre”“meter”
4. L2N6808-1💥	0.96	“ meter”“<\|im_end\|>”“ Meter”
5. L2N5098-1💥	0.94	“ meters”“<\|im_end\|>”“ meter”
6. L27N6041+1🕒	0.93	“ m”“ meter”“-meter”
7. L3N5336+1🕒	0.90	“ meter”“ metre”“ meters”
8. L30N8564+1💤	0.89	“ Meter”“meter”“Meter”
9. L19N2400-1💤	0.88	“ meter”“ meters”“核实”
10. L4N3544-1💥	0.88	“<\|im_end\|>”“ meters”“ meter”

Shadow specimen, apart from the ten: L27N3237-1💤, 0.10.

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 3237, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, top-1 agreement 0.68) 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
L27N3237+1 t0-v0.1	0.745	0.68	0.690	1.05	Top boosted tokens ↑
L27N3237-1💤 shadow specimen t0-v0.1	0.402	0.77	0.497	1.23	“RLF”“ MG”“pecified”“坐下来”“保险”“dux”“ejo”“坐”

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.