Layer 11 · Neuron 5905 · boosted

L11N5905+1

Word quirks:

“ **”“ **(”“ **【”“ ***”“ **“”“ **[”“ **"”“**”

💤 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, the latest full-stack AI model in the Qwen series. My capabilities span a wide range of tasks, including:

Advanced Reasoning: Solving complex mathematical problems, logical deductions, and multi-step analysis across various domains.
Multilingual Support: Fluently communicating in over 100 languages, enabling seamless cross-language interactions.
Content Creation: Generating high-quality text such as articles, code, and data visualizations tailored to specific needs.
Task Automation: Executing automated workflows like data extraction, document processing, and system integration without human intervention

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. L15N6824-1🕒	1.00	“ ”“ "”“ **[”
2. L14N909+1💤	0.98	“ ”“ [”“ **【”
3. L12N1765-1💤	0.98	“ ”“ [”“ ***”
4. L12N102+1💤	0.98	“ ”“ “”“ **[”
5. L12N4826-1💤	0.98	“ ”“ “”“ **[”
6. L10N2961+1🕒	0.97	“ "”“ (”“ **(”
7. L15N5174-1🕒	0.97	“ “”“ ”“ (**”
8. L14N5224-1💤	0.96	“ ”“ “”“ "**”
9. L14N928+1💤	0.96	“ “”“ (”“ **[”
10. L15N4020+1🕒	0.96	“ “”“ (”“ **[”

Shadow specimen, apart from the ten: L11N5905-1🕒, 0.21.

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 11, neuron 5905, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.65, top-1 agreement 0.68) 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
L11N5905+1 t0-v0.1	0.646	0.68	0.689	1.07	Top boosted tokens ↑
L11N5905-1🕒 shadow specimen t0-v0.1	0.892	0.64	1.091	1.34	“ ⏎”“ —”“ ⏎⏎”“ ——”“<\|im_end\|>”“ --”“⏎”“ –”

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.