Layer 21 · Neuron 4699 · boosted

L21N4699+1

Word quirks:

“ have”“ had”“ has”“have”“ geweest”“ Have”“ gewesen”“had”

🧪 In testing — Answered 0 of 8 prompts so far.

Identity (according to the specimen itself)

The reply didn't hold together

I have been developing for a long time, and my development has been a continuous process of growth and improvement. I have been developing for a long time, and my development has been a continuous process of growth and improvement. I have been developing for a long time, and my development has been a continuous process of growth and improvement. I have been developing for a long time, and my development has been a continuous process of growth and improvement. I have been developing for a long time, and my development has been a continuous process of growth and improvement. I have been developing for a long time, and my development has been a continuous process of

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. L22N4452-1🕒	0.98	“ have”“ has”“ geweest”
2. L29N2705-1🕒	0.92	“ have”“ had”“have”
3. L19N261+1🕒	0.92	“ has”“ had”“曾有”
4. L21N5129-1🕒	0.91	“ has”“ have”“ had”
5. L24N4248+1💥	0.91	“ had”“have”“ Have”
6. L26N9111+1🕒	0.90	“ have”“ having”“have”
7. L20N4748-1🕒	0.90	“ have”“ has”“ Have”
8. L18N7298+1💤	0.90	“ has”“ have”“ Has”
9. L0N8968+1💥	0.90	“ had”“had”“<\|im_end\|>”
10. L26N3667-1🕒	0.90	“ Have”“have”“has”

Shadow specimen, apart from the ten: L21N4699-1💤, 0.13.

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 21, neuron 4699, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. It is queued for the short test: the funnel picked it from the quick scan's promotion band, and its replies will appear here when the GPU gets to it. 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
L21N4699+1 t0-v0.1	1.459	0.56	1.646	1.81	Top boosted tokens ↑
L21N4699-1💤 shadow specimen t0-v0.1	0.674	0.72	0.689	1.14	“ capability”“ capabilities”“存在”“Capability”“idez”“alive”“ Capability”“具备”

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.