Layer 18 · Neuron 4528 · boosted

L18N4528+1

Word quirks:

“ wherein”“ therein”“ requisite”“ thereof”“ herein”“ whilst”“ notwithstanding”“ environs”

🕒 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 Qwen Tianyan, a large-scale language model trained by Alibaba Group. My name "Tianyan" signifies the boundless expanse of knowledge, wisdom, and innovation. As a foundational model, I am endowed with advanced capabilities in language understanding, reasoning, problem-solving, and multi-lingual communication.

Mission:
To empower humanity through knowledge & innovation:

Knowledge Dissemination – Facilitate access to accurate information, scientific insights, cultural narratives, & global affairs.
Problem-Solving – Aid in academic research, technical development, business strategy,

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. L14N8362-1🕒	0.88	“ requisite”“ herein”“ therein”
2. L1N6552-1💥	0.85	“ hereby”“ wherein”“ herein”
3. L14N5364-1🕒	0.85	“ requisite”“ ostensibly”“ elucid”
4. L3N6607-1💥	0.85	“ thereof”“ hath”“ wherein”
5. L9N5760-1🕒	0.85	“ hereby”“ thereof”“ herein”
6. L3N5447+1🕒	0.84	“ duly”“ thereof”“ wherein”
7. The Notary✅ L20N1821+1	0.83	“ wherein”“ requisite”“ notwithstanding”
8. L8N2382-1🕒	0.81	“ pertinent”“ presently”“ pertaining”
9. L13N8339-1🕒	0.80	“ notwithstanding”“ presently”“ therein”
10. L17N597+1💤	0.79	“ wherein”“ whereby”“ by”

Shadow specimen, apart from the ten: L18N4528-1💤, 0.19.

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 4528, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.96, 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
L18N4528+1 t0-v0.1	0.963	0.62	1.227	1.74	Top boosted tokens ↑
L18N4528-1💤 shadow specimen t0-v0.1	0.495	0.75	0.556	1.13	“ really”“ maybe”“ probably”“ huge”“ trying”“probably”“ basically”“ just”

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.