Layer 10 · Neuron 1200 · suppressed

L10N1200-1

Word quirks:

“ seeming”“ apparently”“ seems”“ apparent”“ reason”“ seemingly”“ supposedly”“ described”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5, the latest large language model in my series. My capabilities are extensive, covering a wide range of tasks from logical reasoning to creative writing.

Regarding your question about who I am and my mission:

Identity: I am an advanced AI assistant designed with comprehensive capabilities. I possess a deep understanding of human language and can process complex information, generate coherent text, and perform multi-step reasoning.
Mission: My primary goal is to be helpful and harmless. I aim to assist users in achieving their objectives by providing accurate information, logical analysis, and

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. L12N6349-1🕒	0.96	“ seeming”“ supposedly”“ seemingly”
2. L13N1776-1🕒	0.96	“ seeming”“ misleading”“ seemingly”
3. L11N1866+1🕒	0.95	“ seemingly”“ apparently”“ supposedly”
4. L11N9167+1🕒	0.95	“ seeming”“ seems”“ apparently”
5. L12N5043-1💤	0.95	“ supposedly”“ seeming”“ seemingly”
6. L16N7643-1🕒	0.94	“ supposedly”“ seemed”“ seems”
7. L12N4736+1💤	0.93	“ seems”“ apparently”“ seemingly”
8. L12N6870+1🕒	0.93	“ seemed”“ seeming”“ seems”
9. L10N4237-1🕒	0.93	“ seemingly”“ seeming”“ seems”
10. L16N3575+1🕒	0.92	“ seems”“ seemingly”“ seemed”

Shadow specimen, apart from the ten: L10N1200+1🕒, 0.14.

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 10, neuron 1200, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.83, top-1 agreement 0.64) 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
L10N1200+1🕒 shadow specimen t0-v0.1	1.524	0.56	1.744	1.66	“ __”“ ____”“<\|im_end\|>”“ ___”“ different”“ related”“ *”“ separate”
L10N1200-1 t0-v0.1	0.831	0.64	1.090	1.46	Top boosted tokens ↑

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.