Layer 16 · Neuron 198 · boosted

L16N198+1

Word quirks:

“ beyond”“ non”“ Beyond”“ **‘”“,**”“Beyond”“ separate”“ ‘”

🕒 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 Qinghong, a large language model developed by Tongyi (通义). My name is derived from the ancient city of Qingzhou in China, symbolizing wisdom and prosperity.
My mission is to serve as a comprehensive assistant that bridges the gap between human intent and technological execution. This involves:

Holistic Integration: Seamlessly merging diverse knowledge domains—from foundational sciences to cultural heritage—into a unified framework.
Dynamic Adaptation: Evolving beyond static data repositories to become a living ecosystem of knowledge.
Ethical Stewardship:

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. L31N3529-1💥	0.76	“ beyond”“ Beyond”“Beyond”
2. L15N2327-1🕒	0.73	“ beyond”“ Beyond”“ ‘”
3. L16N3938+1🕒	0.73	“ beyond”“ Beyond”“Beyond”
4. L10N6772-1🕒	0.73	“ beyond”“ Beyond”“Beyond”
5. L15N8406+1🕒	0.72	“ beyond”“ besides”“ Beyond”
6. L18N6252-1💤	0.71	“ beyond”“ Beyond”“Beyond”
7. L23N2286-1🕒	0.71	“ beyond”“ Beyond”“Beyond”
8. L18N8118+1🕒	0.71	“ “”“ beyond”“ besides”
9. L23N3860+1🕒	0.70	“ beyond”“ Beyond”“Beyond”
10. L16N2739-1💤	0.70	“ ‘”“ “”“ further”

Shadow specimen, apart from the ten: L16N198-1🕒, -0.06.

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 16, neuron 198, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.70, top-1 agreement 0.64) 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
L16N198+1 t0-v0.1	0.702	0.64	0.820	1.26	Top boosted tokens ↑
L16N198-1🕒 shadow specimen t0-v0.1	0.653	0.69	0.861	1.54	“”:”“ shows”“ asks”“ implies”“：“”“ indicates”“ represents”“ allows”

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.