Layer 13 · Neuron 8613 · suppressed

L13N8613-1

Word quirks:

“ mentioning”“ mentioned”“ mention”“ reference”“ description”“ describing”“ related”“ relevant”

💤 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 with over 100 billion parameters. My capabilities include answering questions, writing code, and performing logical reasoning. I can also handle tasks like data analysis, math problems, and more.

If you'd like to test my knowledge or see how I perform, feel free to ask! Let me know if you want to explore specific features or use cases. 😊

Key Features:

Multilingual Support: I support over 100 languages.
Long Context Understanding: I can process long documents or videos.
**Advanced Reasoning

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. L9N2559-1🕒	0.91	“ mentioning”“ relevant”“ statements”
2. L11N3164-1🕒	0.90	“ mentioned”“ description”“ describing”
3. L7N9013-1💤	0.88	“ referring”“ reference”“ description”
4. L16N6550-1🕒	0.87	“ describing”“ description”“ what”
5. L17N4052-1💤	0.86	“ mentioning”“ mention”“ description”
6. L13N5370+1🕒	0.85	“ typical”“ mentioned”“ description”
7. L11N6000+1🕒	0.85	“ referenced”“ info”“ relevant”
8. L13N1293-1💥	0.84	“ mentioned”“ mention”“ listed”
9. L13N6262+1🕒	0.84	“ description”“ maybe”“ reference”
10. L16N7608+1🕒	0.84	“ description”“ mention”“ describing”

Shadow specimen, apart from the ten: L13N8613+1🕒, 0.16.

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 8613, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, 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
L13N8613+1🕒 shadow specimen t0-v0.1	0.869	0.60	1.162	1.48	“ ——”“ typical”“ —”“ usual”“——”“—you”“ amongst”“ supposedly”
L13N8613-1 t0-v0.1	0.741	0.67	0.913	1.15	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.