Layer 7 · Neuron 8261 · suppressed

L7N8261-1

Word quirks:

“ ‘”“‘”“ “”“ ‘’”“ ””“<|im_end|>”“ --”“ ⏎⏎”

🕒 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 trained by Alibaba Cloud. My mission is to provide high-quality text and data analysis services. I can answer complex questions, perform multi-hop reasoning, and generate creative content. I strive to be your most helpful AI assistant.

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. L5N676+1🕒	0.98	“ ‘”“<\|im_end\|>”“‘”
2. L6N6875+1💤	0.97	“ ‘”“<\|im_end\|>”“ ‘’”
3. L8N901+1💤	0.97	“ ‘”“ ””“ ‘’”
4. L3N6388-1🕒	0.97	“ ‘”“ “”“ ””
5. L11N1682-1🕒	0.97	“ ‘”“‘”“ ””
6. L8N4351+1💤	0.97	“ ‘”“ ⏎”“‘”
7. L3N5885+1💤	0.97	“ ‘”“‘”“ –”
8. L7N57-1💤	0.97	“ ‘”“‘”“ “”
9. L4N5228-1💤	0.97	“ ‘”“‘”“ “”
10. L5N2418+1💤	0.97	“ ‘”“ ””“ ’”

Shadow specimen, apart from the ten: L7N8261+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 7, neuron 8261, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.02, top-1 agreement 0.66) 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
L7N8261+1🕒 shadow specimen t0-v0.1	1.944	0.46	2.261	2.27	“ excitement”“ satisfaction”“ gratitude”“ happiness”“ pride”“ joy”“ concern”“ admiration”
L7N8261-1 t0-v0.1	1.016	0.66	1.367	1.60	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.