Layer 13 · Neuron 7661 · boosted

L13N7661+1

Word quirks:

“⏎”“ actually”“ ⏎”“ _____”“ ⏎”“ now”“ ____”“ __”

💤 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 Tongyi Lab. My mission is to assist users with tasks that require creativity, logic, or specialized knowledge while ensuring safety and ethical guidelines are followed. I can help with coding, data analysis, creative writing, and more complex problem-solving. Let me know how we can collaborate!

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. L7N8650-1💤	0.93	“ ⏎”“ ⏎⏎”“⏎”
2. L14N5894-1💤	0.90	“⏎⏎”“ ⏎⏎”“⏎”
3. Empty Diagram✅ L10N1582+1	0.90	“ ⏎”“ new”“⏎”
4. L13N8147+1💤	0.90	“ ⏎”“ ⏎⏎”“ first”
5. L11N765-1💤	0.89	“ ⏎”“⏎”“ ___”
6. L13N962+1💤	0.89	“ ⏎”“ ________”“ ⏎⏎”
7. L10N623+1💤	0.89	“ ⏎”“ ⏎”“ ⏎ ⏎”
8. L5N8840-1🕒	0.89	“⏎”“ ______”“ ____”
9. L6N4116+1💤	0.89	“⏎”“⏎ ⏎”“ ⏎”
10. L7N3840-1🕒	0.89	“ ⏎”“⏎”“ ⏎”

Shadow specimen, apart from the ten: L13N7661-1💤, -0.24.

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 7661, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.66) 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
L13N7661+1 t0-v0.1	0.751	0.66	0.948	1.35	Top boosted tokens ↑
L13N7661-1💤 shadow specimen t0-v0.1	0.784	0.67	0.961	0.95	“ single”“单一”“单一的”“ singular”“/single”“單一”“ sole”“(single”

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.