Layer 10 · Neuron 7437 · suppressed

L10N7437-1

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, a large language model developed by Tongyi Lab. I can understand and generate text in any language, including Chinese for Chinese users and English for English users. My capabilities include:

Text Generation: I can generate coherent text based on user input. For example, if you ask me to write a poem, I will generate a poem in the specified style. If you ask me to write code, I will generate code in the specified programming language.
Dialogue Understanding: I can understand the context of your dialogue and respond appropriately. For example, if you ask me about

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. L14N7676+1🕒	0.95	“ "”“ __”“ *”
2. L15N3380+1💤	0.94	“ **"”“ __”“ ____”
3. L7N3911-1💤	0.94	“ ____”“ ______”“ ***”
4. L10N2107-1💤	0.94	“ **“”“ ____”“ __”
5. L10N6444+1🕒	0.94	“ *”“ ”“ ______”
6. L6N3816+1💤	0.93	“⏎”“ *”“ ”
7. L6N5835-1💤	0.93	“⏎”“ _____”“ **"”
8. L15N6905+1💤	0.93	“ **”“ __”“ ”
9. L16N7631+1🕒	0.93	“ "”“ ____”“ *”
10. L12N8612-1🕒	0.93	“ **”“ ____”“ ”

Shadow specimen, apart from the ten: L10N7437+1🕒, 0.19.

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 7437, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.86, 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
L10N7437+1🕒 shadow specimen t0-v0.1	1.518	0.52	1.804	1.74	“<\|im_end\|>”“ bigger”“ noticeable”“ “”“ notable”“ atleast”“ ‘”“ enough”
L10N7437-1 t0-v0.1	0.857	0.66	1.096	1.32	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.