Layer 16 · Neuron 2885 · suppressed

L16N2885-1

Word quirks:

“ recently”“ currently”“ recent”“ interesting”“ presently”“ now”“ fascinating”“ lately”

🕒 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 a large language model created by Alibaba Cloud, and I have been designed to be a versatile and helpful assistant with a wide range of capabilities. My mission is to provide high-quality assistance across various domains, including answering questions, solving problems, creating content, and more. I aim to be an intelligent and efficient partner that has made significant progress in recent years. If you have any specific tasks or questions, 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. L12N3949-1🕒	0.92	“ currently”“ recently”“ now”
2. L10N8526-1🕒	0.92	“ recently”“ now”“ similar”
3. L11N1568+1💤	0.92	“ currently”“ presently”“ recently”
4. L11N7646+1🕒	0.90	“ recently”“ already”“ currently”
5. L12N4932+1🕒	0.89	“ current”“ recent”“ now”
6. L15N3637-1🕒	0.88	“ currently”“ presently”“ today”
7. L9N2253-1💤	0.88	“ recently”“ lately”“ recent”
8. L14N7627-1🕒	0.86	“ recently”“ currently”“ nicer”
9. L13N1334-1💤	0.86	“ now”“<\|im_end\|>”“ nowadays”
10. L18N2714+1💤	0.86	“ recently”“ recent”“ currently”

Shadow specimen, apart from the ten: L16N2885+1🕒, 0.27.

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 2885, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.63, top-1 agreement 0.72) 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
L16N2885+1🕒 shadow specimen t0-v0.1	0.887	0.64	1.051	1.28	“ either”“ whether”“ whatever”“ usually”“ if”“ optionally”“ used”“ possibly”
L16N2885-1 t0-v0.1	0.635	0.72	0.837	1.27	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.