Layer 18 · Neuron 1528 · suppressed

L18N1528-1

Word quirks:

“ listed”“ named”“ called”“ described”“ included”“ including”“ includes”“ include”

🕒 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, an AI named after the large model developed by Alibaba Cloud. I am a large language model that specializes in understanding and processing complex tasks across various domains, including:

Comprehensive capabilities:

AI
Natural Language Processing (NLP): Including text generation, such as creative writing like poetry, stories, and in different styles.
Code for both text and code: Ranging from programming tasks like generating code for web projects like "a simple Python script to analyze data" to more complex ones like "a full-stack application to build a multi

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. L15N8370-1🕒	0.83	“ listed”“ included”“ of”
2. L13N7128-1🕒	0.83	“ included”“ specific”“ including”
3. L17N7674+1🕒	0.83	“ included”“included”“ mentioned”
4. L15N3072-1🕒	0.83	“ including”“ include”“ includes”
5. L14N4714-1🕒	0.83	“ includes”“ include”“ including”
6. The Namer✅ L17N2222+1	0.82	“叫做”“ called”“ includes”
7. L15N4660+1🕒	0.82	“ includes”“ modified”“ special”
8. L15N5961-1🕒	0.82	“⏎”“ included”“ include”
9. L12N8966-1🕒	0.81	“ included”“ includes”“ include”
10. L20N7492-1🕒	0.81	“ included”“ Included”“included”

Shadow specimen, apart from the ten: L18N1528+1🕒, 0.11.

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 1528, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.92, 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
L18N1528+1🕒 shadow specimen t0-v0.1	0.774	0.64	0.868	1.24	“ and”“/”“/F”“/Sh”“-centric”“-focused”“/C”“/native”
L18N1528-1 t0-v0.1	0.915	0.62	1.032	1.50	Top boosted tokens ↑

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.