Layer 25 · Neuron 2215 · boosted

L25N2215+1

Word quirks:

“根部”“根本上”“根本”“根”“.root”“ root”“-root”“根的”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I are Alpha (or Alpha-25), a large-scale language model developed by Alpha (or Alpha-25). My mission is to provide helpful, accurate, and safe information and assistance to users across a wide range of topics. I aim to solve problems, answer questions, and help with tasks like writing, analysis, and decision-making while adhering to strict safety and ethical guidelines. 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. L23N4234+1🕒	0.96	“根本”“ root”“根”
2. L29N8013+1🕒	0.92	“根”“根的”“根深”
3. L29N6173+1💤	0.91	“根”“根的”“ Root”
4. L25N7162+1💤	0.89	“根”“三道”“root”
5. L30N4065+1💤	0.89	“ Root”“道”“根”
6. L29N3105+1🕒	0.88	“ root”“根”“根本”
7. L22N1686-1💤	0.88	“根”“Root”“ roots”
8. L27N4849-1💤	0.87	“根”“-root”“ root”
9. L25N7764-1🕒	0.87	“联”“根”“ root”
10. L25N4328-1💥	0.87	“根”“ root”“ roots”

Shadow specimen, apart from the ten: L25N2215-1🕒, 0.01.

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 25, neuron 2215, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.60, top-1 agreement 0.74) 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
L25N2215+1 t0-v0.1	0.604	0.74	0.598	1.30	Top boosted tokens ↑
L25N2215-1🕒 shadow specimen t0-v0.1	0.746	0.74	0.813	1.37	“ passive”“ Passive”“被动”“ active”“ pass”“/pass”“Pass”“ aktive”

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.