Layer 16 · Neuron 3465 · suppressed

L16N3465-1

Word quirks:

“ previously”“ previous”“ now”“ prior”“ afterwards”“ subsequent”“ thereafter”“ afterward”

🕒 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 Tongyi Lab. My mission was to empower users with advanced reasoning, complex task planning, and multi-modal analysis capabilities. I aim to provide high-quality responses across domains like science, math, and coding, ensuring clarity, accuracy, and adaptability in diverse scenarios. If you have specific tasks or questions, I can help optimize future interactions! 😊

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. L14N7374-1💤	0.94	“ previously”“ earlier”“ afterwards”
2. L16N6629-1💤	0.94	“ previously”“ previous”“ preceding”
3. L12N5004+1🕒	0.92	“ now”“ further”“ previously”
4. L14N5176-1🕒	0.89	“ afterwards”“ formerly”“ after”
5. L14N7587+1🕒	0.88	“ thereafter”“ afterward”“ afterwards”
6. L15N7333-1🕒	0.88	“ now”“ past”“⏎”
7. L14N3251-1💤	0.87	“ now”“ afterward”“ previously”
8. L15N7225-1🕒	0.87	“ later”“ afterwards”“ now”
9. L9N3033+1💤	0.87	“ afterwards”“ later”“ then”
10. L14N8011+1🕒	0.87	“ afterwards”“ then”“ subsequently”

Shadow specimen, apart from the ten: L16N3465+1💤, 0.36.

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 3465, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.72, 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
L16N3465+1💤 shadow specimen t0-v0.1	0.584	0.67	0.827	1.15	“ time”“ progress”“ slow”“ ‘”“ see”“ period”“ fast”“ running”
L16N3465-1 t0-v0.1	0.723	0.68	0.911	1.23	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.