Layer 16 · Neuron 8494 · suppressed

L16N8494-1

Word quirks:

“ rewritten”“ transformed”“ converted”“ changed”“ concept”“ rewrite”“ conversion”“ created”

💤 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 trained by Alibaba Cloud. I am designed to assist users in completing tasks through understanding, learning, and generating text, code, and more. My core capabilities include:

Advanced Context Understanding: I can process and analyze complex tasks with enhanced precision, supporting fluent interaction even in nuanced scenarios.
Multilingual Support: I support over 100 languages, enabling me to assist users globally.
Specialized Capabilities:

Full-stack development: From coding to deployment.
Visual analysis: Interpreting

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. L18N5376-1🕒	0.91	“ converted”“ transformed”“ renamed”
2. L12N7254+1🕒	0.85	“ converted”“ changed”“ changing”
3. L17N8901+1🕒	0.84	“ converted”“ transformed”“ modified”
4. L7N6429+1🕒	0.84	“ converted”“ conversion”“ conversions”
5. L15N7258-1🕒	0.84	“ changed”“ converted”“ replaced”
6. Great News✅ L12N7152-1	0.83	“ now”“ changed”“ change”
7. L14N7727+1💤	0.82	“ changed”“ transforming”“ changing”
8. L6N7870+1🕒	0.80	“ transformation”“ transformed”“ transformations”
9. L9N5198-1🕒	0.80	“ changed”“ replaced”“ modified”
10. L12N7709+1🕒	0.80	“ transformed”“ related”“ transformation”

Shadow specimen, apart from the ten: L16N8494+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 16, neuron 8494, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.45, top-1 agreement 0.77) 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
L16N8494+1🕒 shadow specimen t0-v0.1	1.042	0.61	1.327	1.55	“ correctly”“ properly”“ correct”“ proper”“ appropriately”“ consistent”“ conforms”“ *[”
L16N8494-1 t0-v0.1	0.450	0.77	0.564	1.08	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.