Layer 13 · Neuron 7005 · boosted

L13N7005+1

Word quirks:

“ now”“ updated”“ newer”“ bigger”“ new”“ \"”“ NOW”“ changed”

🕒 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 in the Qwen series. My capabilities include advanced reasoning, visual analysis, and full-stack coding support. I can help with complex tasks like generating code, analyzing charts, and handling multi-step workflows. If you have specific questions or need assistance, feel free to let me know!

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. L11N3113+1🕒	0.91	“ now”“ newer”“ current”
2. L12N2263+1🕒	0.86	“ current”“ now”“ newer”
3. L11N6695+1🕒	0.85	“ newer”“ updated”“ modern”
4. L10N6591+1🕒	0.85	“ new”“ now”“ needed”
5. The Lockdown Diarist✅ L9N3698-1	0.84	“ now”“ really”“ quicker”
6. L16N5281+1🕒	0.83	“ now”“ changed”“ newer”
7. L12N9000-1🕒	0.83	“ now”“ updated”“ newer”
8. L9N4757+1🕒	0.82	“ new”“ update”“ updated”
9. L11N721+1🕒	0.82	“ now”“ updated”“ changed”
10. L11N3459-1💤	0.82	“ now”“ new”“ NEW”

Shadow specimen, apart from the ten: L13N7005-1💤, -0.04.

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 7005, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, 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
L13N7005+1 t0-v0.1	0.812	0.64	1.024	1.28	Top boosted tokens ↑
L13N7005-1💤 shadow specimen t0-v0.1	0.773	0.67	1.006	1.14	“ ⏎”“ ⏎”“ ⏎”“！（”“ ⏎⏎”“ ”“ "”“!”

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.