Layer 13 · Neuron 60 · suppressed

L13N60-1

Word quirks:

“ --”“ ——”“ –”“ -”“ —”“ ---”“ basically”“...*”

🕒 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)

I am Qwen3.5, a large language model developed by the Alibaba Group. I'm not just another chatbot—I'm designed to be a versatile assistant for real-world tasks like coding, data analysis, or creative writing. Think of me as your go-to companion for everything from homework to business strategy! 😊

(Note: This is a self-introduction by the user, not a system message.)
</think>

I am Qwen3.5, a large language model developed by Tongyi Lab. I'd love to help you with this! Let's dive in together. 😊
</think>

I am

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. L11N326+1🕒	0.95	“ --”“ –”“ ——”
2. L18N631-1💤	0.95	“ —”“ –”“ ——”
3. L16N1925+1💤	0.95	“ --”“ —”“ ——”
4. L23N4995-1⛔	0.94	“ –”“ —”“ ——”
5. L3N8373-1🕒	0.94	“ --”“ ——”“ —”
6. L12N1200-1💤	0.94	“ --”“ —”“—but”
7. L16N6182-1🕒	0.94	“ —”“ ——”“ --”
8. L17N3167-1💤	0.94	“ —”“ ——”“ -”
9. L14N507-1🕒	0.94	“ —”“ --”“ ...”
10. L17N213-1💤	0.94	“ —”“ --”“ ——”

Shadow specimen, apart from the ten: L13N60+1💤, 0.16.

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 60, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.84, 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
L13N60+1💤 shadow specimen t0-v0.1	0.652	0.68	0.844	1.16	“⏎”“ ⏎”“⏎⏎”“ ⏎ ⏎”“ ⏎”“ **”“⏎ ⏎”“ ⏎⏎”
L13N60-1 t0-v0.1	0.843	0.64	0.999	1.44	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.