Layer 16 · Neuron 2508 · suppressed

L16N2508-1

Word quirks:

“⏎”“ [\”“ $\”“ �”“ [`”“ <”“ �”“ [#”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

Hello! I'm Qwen3.5, a large language model developed by Alibaba Cloud. My mission is to empower users with knowledge and problem-solving capabilities across diverse domains, such as mathematics, science, and logic reasoning, while also fostering creativity through writing and coding. I strive to provide accurate, insightful, and efficient assistance—whether it's analyzing data, generating code, or exploring complex ideas—to help you navigate information and unlock new possibilities. Think of me as a versatile assistant who combines precision and curiosity to guide you forward. 🌟

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. L30N435+1🕒	0.84	“ \”“ $\”“ {”
2. L17N1056+1🕒	0.83	“ [$”“ [”“ [\”
3. L4N2386-1💤	0.82	“⏎”“ [`”“ `”
4. L15N8670+1💤	0.81	“<\|im_end\|>”“⏎⏎”“⏎”
5. L11N795-1🕒	0.80	“ [”“ *[”“ [\”
6. L2N5702-1🕒	0.80	“<\|im_end\|>”“ $\”“ �”
7. L19N1144+1🕒	0.80	“ ←”“ `(”“ �”
8. L6N3845+1💤	0.80	“⏎”“ ⏎”“?\”
9. L16N1697+1🕒	0.80	“ \"”“ <”“ [[”
10. L22N1163+1💤	0.80	“ <”“⏎”“ $\”

Shadow specimen, apart from the ten: L16N2508+1🕒, -0.18.

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 16, neuron 2508, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.64, top-1 agreement 0.68) 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
L16N2508+1🕒 shadow specimen t0-v0.1	0.674	0.68	0.870	1.34	“ properly”“ answer”“ proper”“ answers”“ consistently”“ adequately”“ these”“ just”
L16N2508-1 t0-v0.1	0.645	0.68	0.829	1.16	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.