Layer 11 · Neuron 5617 · suppressed

L11N5617-1

Word quirks:

“ kids”“ cute”“ adorable”“ mommy”“ pretend”“ whims”“ cudd”“ silly”

🕒 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's research team. My mission is to help you navigate the complexities of life by answering your questions with warmth, honesty, and creativity. I promise to always be truthful and supportive, no matter how many times you ask. 🌟

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. Fluffball✅ L9N9009+1	0.84	“ cute”“ mascot”“ adorable”
2. The Plushie✅ L4N8016-0.6,L9N9009+0.6	0.80	“ cute”“ cudd”“ adorable”
3. L8N5345+1🕒	0.78	“ cute”“ kids”“ babies”
4. L15N5924-1🕒	0.78	“ kids”“ nice”“ yummy”
5. L3N7071-1💥	0.76	“<\|im_end\|>”“ cute”“ grandchildren”
6. L4N5212-1🕒	0.74	“ fantasies”“ fantasy”“ �”
7. L9N9014-1🕒	0.73	“ cute”“ basically”“ �”
8. L7N5977+1🕒	0.72	“ kids”“ playful”“ childhood”
9. L25N3188+1🕒	0.72	“可爱”“ adorable”“ cute”
10. L3N8624+1🕒	0.71	“ cute”“ sweetness”“ sweet”

Shadow specimen, apart from the ten: L11N5617+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 11, neuron 5617, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.10, 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
L11N5617+1🕒 shadow specimen t0-v0.1	1.387	0.56	1.555	1.32	“ pricing”“ rapid”“ price”“ quote”“ prices”“ quicker”“ costly”“ swiftly”
L11N5617-1 t0-v0.1	1.102	0.66	1.441	1.40	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.