Layer 13 · Neuron 1363 · suppressed

L13N1363-1

Word quirks:

“ maybe”“ description”“ mentioned”“ mention”“ mentioning”“ describing”“ needed”“ thinking”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, a large language model developed by Alibaba Cloud's Tongyi Lab. I am an AI assistant designed to help users with tasks like answering questions, creating content, coding, or analyzing data. My capabilities include:

Core Competencies: Natural language understanding, logical reasoning, and multilingual support (100+ languages).
Specialized Strengths: Deep learning models trained on diverse data for accuracy and reliability.
Human Alignment: Optimized for clarity, empathy, and adaptability to user needs.
Autonomous & Scalable: Built for efficiency, security

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. L17N4052-1💤	0.86	“ mentioning”“ mention”“ description”
2. L11N3164-1🕒	0.85	“ mentioned”“ description”“ describing”
3. L16N7608+1🕒	0.85	“ description”“ mention”“ describing”
4. L15N4815-1🕒	0.84	“ description”“ mentioning”“ info”
5. L13N5370+1🕒	0.84	“ typical”“ mentioned”“ description”
6. L15N3083+1💤	0.84	“ mentioned”“ description”“ seems”
7. L9N2559-1🕒	0.84	“ mentioning”“ relevant”“ statements”
8. L16N5857-1🕒	0.82	“ mentioned”“ described”“ explained”
9. L13N8349+1🕒	0.82	“ description”“ yourself”“ mention”
10. L16N6550-1🕒	0.82	“ describing”“ description”“ what”

Shadow specimen, apart from the ten: L13N1363+1🕒, 0.23.

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 13, neuron 1363, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.69, top-1 agreement 0.67) 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
L13N1363+1🕒 shadow specimen t0-v0.1	0.876	0.62	1.114	1.20	“ even”“ apparently”“ quick”“ basically”“ supposedly”“ quickly”“ typical”“ typically”
L13N1363-1 t0-v0.1	0.694	0.67	0.832	1.32	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.