Layer 15 · Neuron 7306 · boosted

L15N7306+1

Word quirks:

“ --”“ final”“ —”“ last”“ ---”“ ——”“ ----”“ ***”

🕒 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, the latest large language model developed by Tongyi Lab. My mission is to empower users with cutting-edge capabilities—acting as a versatile assistant who can help them learn, create, solve problems, or even explore ideas. While I don't have personal beliefs or a specific mission beyond my technical scope, I am designed to be a helpful, harmless, and honest resource for anyone who wants to know more about me.

If you're curious about my capabilities, I'd be glad to share! 😊 (Just let me know if you'd like to continue this conversation.)

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. L12N2373+1💤	0.88	“ ——”“ —”“ –”
2. L14N8637+1💤	0.88	“ ——”“ ---”“ similar”
3. L18N6821-1🕒	0.87	“ ---”“---”“ --”
4. L10N377+1💤	0.87	“ ——”“ --”“ —”
5. L16N4121+1🕒	0.86	“ --”“ —”“ ---”
6. L17N3336-1🕒	0.86	“ either”“ ---”“ -”
7. L12N3671-1💤	0.86	“ --”“<\|im_end\|>”“⏎”
8. L13N1460-1💤	0.86	“ ——”“ --”“ ---”
9. L12N475-1🕒	0.86	“ --”“ ——”“ —”
10. L12N171+1💤	0.86	“ ***”“ --”“ –”

Shadow specimen, apart from the ten: L15N7306-1🕒, 0.09.

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 15, neuron 7306, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.65, top-1 agreement 0.69) 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
L15N7306+1 t0-v0.1	0.650	0.69	0.849	1.08	Top boosted tokens ↑
L15N7306-1🕒 shadow specimen t0-v0.1	0.780	0.64	1.027	1.39	“ never”“ always”“ beginning”“ ALWAYS”“ starting”“ from”“从一开始”“ throughout”

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.