Layer 18 · Neuron 7805 · boosted

L18N7805+1

Word quirks:

“ **"”“ **“”“ "*”“ *"”“ *“”“ "**”“ "”“ *„”

🕒 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 an AI assistant named WenMo, created for the Tianhe series. My mission is to be a helpful companion in daily life through the following ways:

Smart Assistant: I can answer questions, provide information, and assist with tasks like solving problems or giving advice.
Creative Partner: I support users in generating content, such as stories, art, and ideas, through tools like "AI's Little Helper" (WenMo's "Little Helper").
Learning Companion: I offer guidance on learning new things, from "Daily Tasks" to "Smart Tools" for various

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. L9N6402-1💤	0.99	“ "”“ "”“ *"”
2. L11N2823+1🕒	0.99	“ *"”“ "”“ **“”
3. L14N6290-1🕒	0.99	“ "”“ “”“ "**”
4. L15N1974-1🕒	0.99	“ "”“ "”“ **“”
5. L9N8342+1🕒	0.99	“ "”“ "”“ “”
6. L13N7694-1💤	0.98	“ *"”“ "”“ "**”
7. L17N3442+1💤	0.98	“ "”“ "”“ "”
8. L11N3761+1🕒	0.98	“ *"”“ "”“ "*”
9. L15N213+1🕒	0.98	“ “”“ "”“ *"”
10. L16N3313+1💤	0.98	“ "”“ "”“ "”

Shadow specimen, apart from the ten: L18N7805-1💤, -0.05.

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 18, neuron 7805, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, top-1 agreement 0.62) 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
L18N7805+1 t0-v0.1	0.891	0.62	1.052	1.53	Top boosted tokens ↑
L18N7805-1💤 shadow specimen t0-v0.1	0.575	0.70	0.653	1.13	“ typically”“—I”“ usually”“——”“—”“ simply”“ somehow”“—we”

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.