Layer 30 · Neuron 5879 · boosted

L30N5879+1

Word quirks:

“陈”“-c”“ C”“.c”“C”“)c”“ c”“ getC”

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

Identity (according to the specimen itself)

The reply didn't hold together

C-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c-c

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. L24N6501+1🕒	0.97	“ c”“ C”“_c”
2. L30N6751+1🕒	0.96	“c”“ c”“陈”
3. L28N7342-1🕒	0.95	“*c”“ c”“-c”
4. L26N4189+1🕒	0.95	“ c”“陈”“ C”
5. L28N8890-1🕒	0.95	“ c”“:c”“陈”
6. L28N8343-1🕒	0.95	“ c”“c”“陈”
7. L28N3817+1🕒	0.95	“ c”“陈”“ C”
8. L28N7435+1🕒	0.95	“ c”“陈”“"c”
9. L30N5314-1🕒	0.94	“C”“ c”“ C”
10. L27N8635-1🕒	0.94	“ c”“$c”“c”

Shadow specimen, apart from the ten: L30N5879-1🕒, 0.38.

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 30, neuron 5879, 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
L30N5879+1 t0-v0.1	1.094	0.62	1.137	1.08	Top boosted tokens ↑
L30N5879-1🕒 shadow specimen t0-v0.1	0.749	0.71	0.829	1.72	“acos”“aco”“cos”“ cord”“ cos”“ Cos”“ cob”“Co”

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.