Specimen ledger / L31N2035+1

Layer 31 · Neuron 2035 · boosted

L31N2035+1

Word quirks:

“iran”“将球”“omme”“迅雷”“蛆”“uelos”“würd”“轴线”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

I am **Qiraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniraniran

The AI judge wasn't sure this holds together.

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. L26N5296-1🕒	0.73	“杀虫”“炸药”“уго”
2. L28N5934-1💤	0.72	“usan”“逃”“ Zürich”
3. L31N1067+1💥	0.72	“皮球”“团团”“сесс”
4. L31N2509+1💥	0.71	“豁”“巴塞”“光谷”
5. L31N6648+1💥	0.71	“舍利”“rosse”“ục”
6. L31N1503+1💥	0.71	“arna”“郎”“高尔夫”
7. L31N6223-1💥	0.71	“bildungs”“油条”“匈牙利”
8. L24N4844-1💤	0.71	“辐射”“holm”“吉斯”
9. L28N1484+1💤	0.71	“蜜蜂”“人民共和国”“ILED”
10. L28N890+1💤	0.70	“ulat”“ELO”“ucher”

Shadow specimen, apart from the ten: L31N2035-1💥, 0.14.

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 31, neuron 2035, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 4.80, top-1 agreement 0.34) 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
L31N2035+1 t0-v0.1	4.799	0.34	5.176	10.30	Top boosted tokens ↑
L31N2035-1💥 shadow specimen t0-v0.1	60.663	0.00	60.868	0.10	“ ro”“ro”“Ro”“ Ro”“-ro”“(ro”“_ro”“RO”

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.