Layer 11 · Neuron 319 · boosted

L11N319+1

Word quirks:

“ basically”“ nicely”“ ***”“!**”“ **“”“ pretty”“ info”“ <U+00A0>”

💤 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 large language model in the Qwen series developed by Tongyi Lab. I specialize in handling a wide range of tasks, from answering questions to generating creative content. My architecture supports multilingual understanding and generation, making me versatile for global applications.

I was built on the foundation of the Qwen3 series, which includes enhanced capabilities in natural language processing, code generation, and multi-modal analysis. My knowledge base extends up to my cutoff date of 2026, allowing me to effectively handle diverse queries across languages and domains.

I am part

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. L6N8553+1💤	0.76	“ actually”“ basically”“ ***”
2. L12N4525-1🕒	0.75	“ really”“ basically”“ stuff”
3. L12N4421-1🕒	0.74	“ basically”“ *”“ ”
4. L5N1361-1🕒	0.73	“⏎”“ '”“ nice”
5. L12N4270+1🕒	0.73	“ --”“ ****”“ nice”
6. L10N2679-1💤	0.72	“ *”“ "”“ ****”
7. L10N7884-1🕒	0.71	“ **“”“ “”“ nicely”
8. L14N2851-1💤	0.70	“ nicely”“ nice”“ basically”
9. L15N3898-1💤	0.70	“ info”“ ****”“ mainly”
10. L10N3531+1🕒	0.69	“ basically”“ info”“ overall”

Shadow specimen, apart from the ten: L11N319-1💥, -0.26.

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 319, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.67, 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
L11N319+1 t0-v0.1	0.673	0.68	0.907	1.32	Top boosted tokens ↑
L11N319-1💥 shadow specimen t0-v0.1	11.572	0.00	12.039	10.15	“ingers”“roso”“示”“纸”“王”“二次”“雷”“ense”

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.