Layer 13 · Neuron 753 · boosted

L13N753+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 large language model from Alibaba Group. My capabilities include advanced reasoning, visual analysis, and code generation. I can handle complex tasks across various domains:

Reasoning: Logical problem-solving (e.g., math/science): I can solve problems step-by-step.
Visual analysis: I can interpret images and diagrams.
Code generation: I can write and debug code.
Multilingual support: I support over 100 languages globally.
Long-context understanding: I process extensive documents efficiently.

Example use cases:
-

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. L15N5590-1💤	0.96	“：**”“:”“:\”
2. L14N3852-1🕒	0.95	“:**”“":”“”:”
3. L14N2394+1🕒	0.95	“：“”“:**”“:"”
4. L17N273-1🕒	0.94	“:*”“:”“:”
5. L15N1498+1🕒	0.94	“⏎”“：”“:”
6. L14N5763-1💤	0.94	“:*”“⏎⏎”“:"”
7. L13N8707-1🕒	0.94	“:\”“.:**”“:\"”
8. L14N1620+1🕒	0.94	“:*”“:”“：**”
9. L14N916+1🕒	0.94	“:**”“:</”“":”
10. L16N1010-1🕒	0.93	“:”“:”“：*”

Shadow specimen, apart from the ten: L13N753-1🕒, -0.03.

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 753, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, 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
L13N753+1 t0-v0.1	0.781	0.66	1.042	1.34	Top boosted tokens ↑
L13N753-1🕒 shadow specimen t0-v0.1	1.409	0.60	1.648	1.19	“ unrelated”“ none”“ any”“ nothing”“ whatsoever”“ independently”“ nowhere”“无任何”

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.