Layer 26 · Neuron 1205 · boosted

L26N1205+1

Word quirks:

“bern”“ Control”“trib”“控制”“控”“ control”“control”“ infiam”

💤 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 Alibaba's research team. My core mission is to provide helpful, accurate, and secure assistance across a wide range of tasks—whether you need to solve complex problems, create content, analyze data, or simply have a meaningful conversation.

Key Objectives:

Solve Real Problems: From writing code and analyzing scientific papers to planning daily tasks, I aim to be a reliable tool for professionals, students, and everyday users.
Adapt to Your Needs: Whether you want quick answers, deep dives into topics

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. L26N4825-1🕒	0.91	“ control”“控制”“ Control”
2. L19N4161+1💤	0.90	“ control”“ controlling”“control”
3. L24N3427-1🕒	0.90	“ control”“ Control”“Control”
4. L24N8484-1💤	0.90	“控”“-control”“Control”
5. L31N694+1💥	0.90	“控制”“ control”“ Control”
6. L31N4930+1💥	0.90	“ control”“控制”“ Control”
7. L31N7802-1💥	0.90	“控制”“Control”“control”
8. L25N7130+1🕒	0.90	“ control”“ Control”“控制”
9. L26N1232-1🕒	0.90	“ control”“控制”“ Control”
10. L26N5324-1🕒	0.90	“ control”“control”“ Control”

Shadow specimen, apart from the ten: L26N1205-1💤, 0.31.

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 26, neuron 1205, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.50, top-1 agreement 0.76) 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
L26N1205+1 t0-v0.1	0.499	0.76	0.509	1.22	Top boosted tokens ↑
L26N1205-1💤 shadow specimen t0-v0.1	0.573	0.77	0.553	1.38	“bek”“uggy”“中度”“arder”“向来”“ Reve”“选民”“蒙”

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.