Layer 12 · Neuron 4181 · boosted

L12N4181+1

Word quirks:

“ here”“?:”“ now”“ [”“ "[”“ "...”“:"”“:”

🕒 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 am Qwen3.5, a large language model with a 256K context window. My mission is to empower users by providing high-quality, accurate, and unbiased information. I don't want to be the source of misinformation or bias in my outputs.

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. The Ventriloquist's Dummy✅ L12N1733+1	0.78	“<\|im_end\|>”“ [”“ ____”
2. L6N3842-1🕒	0.72	“⏎”“ [”“ ⏎”
3. L13N795+1💤	0.71	“ [”“ now”“.[”
4. L13N3178-1💤	0.71	“:"”“ here”“:”
5. L13N88+1💤	0.70	“:\”“ [”“:*”
6. L16N4032-1🕒	0.70	“:”“:\”“:...”
7. L11N4165-1🕒	0.69	“?:”“⏎”“<\|im_end\|>”
8. L11N2695-1💤	0.69	“:"”“:\”“.:”
9. L17N1882-1💤	0.69	“ [”“?[”“_[”
10. L13N3652-1🕒	0.69	“ here”“:”“**:”

Shadow specimen, apart from the ten: L12N4181-1🕒, -0.04.

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 12, neuron 4181, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.94, top-1 agreement 0.63) 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
L12N4181+1 t0-v0.1	0.939	0.63	1.203	1.47	Top boosted tokens ↑
L12N4181-1🕒 shadow specimen t0-v0.1	1.079	0.63	1.285	1.33	“ simple”“ specific”“ usually”“ sometimes”“ isn”“ simpler”“ typically”“ either”

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.