Layer 17 · Neuron 8957 · suppressed

L17N8957-1

Word quirks:

“ describing”“ description”“ implicit”“ describe”“ descriptions”“ unnamed”“ name”“ unclear”

🕒 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 Qwen3.1, a large language model developed by Tongyi Lab. As an AI assistant, I don't have a personal identity or a "mission" in the way humans do—I don't need to help myself achieve anything. However, my purpose is clear: to empower users by providing intelligent, creative, and actionable support across a wide range of tasks.

How I Help:

For creators: Generate text, code, and ideas to spark innovation.
For professionals: Deliver precise analysis, problem-solving, and multi-step reasoning for complex queries.
**For everyday

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. L14N7649+1🕒	0.85	“ description”“ descriptions”“ describing”
2. L12N7584+1🕒	0.85	“ description”“ descriptions”“ describing”
3. L10N1740-1🕒	0.84	“ description”“ descriptions”“ described”
4. L18N7342+1🕒	0.84	“ describing”“ description”“ descriptions”
5. L17N5489-1🕒	0.84	“ description”“ described”“ describing”
6. L9N8019+1💤	0.84	“ description”“ describing”“ descriptions”
7. L26N3747+1🕒	0.84	“ Description”“Description”“ description”
8. L11N8503+1🕒	0.84	“ description”“ describing”“ only”
9. L13N5548-1🕒	0.83	“ description”“ details”“ describes”
10. L10N7332-1🕒	0.82	“ description”“ descriptions”“ describing”

Shadow specimen, apart from the ten: L17N8957+1🕒, 0.07.

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 17, neuron 8957, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, top-1 agreement 0.71) 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
L17N8957+1🕒 shadow specimen t0-v0.1	0.830	0.64	0.928	1.32	“ optimal”“最优”“ optimum”“ best”“ rather”“ suitability”“ strongest”“ whichever”
L17N8957-1 t0-v0.1	0.732	0.71	0.892	1.18	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.