Layer 13 · Neuron 8570 · suppressed

L13N8570-1

Word quirks:

“ unexpected”“ unexpectedly”“ new”“ suddenly”“ unfore”“ seems”“ somehow”“ discovery”

🕒 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, the latest large language model developed by Alibaba Cloud's Tongyi Lab. My capabilities include advanced reasoning, creative writing, and deep analysis across various domains. I am designed to assist users in solving complex problems with precision and clarity.

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. L6N6763+1🕒	0.87	“ unexpected”“ unsus”“ sudden”
2. L8N8696-1🕒	0.86	“ unexpected”“ sudden”“ unexpectedly”
3. L15N6286+1🕒	0.85	“ unexpected”“ unexpectedly”“ somehow”
4. L13N8331+1🕒	0.83	“ suddenly”“ sometimes”“ sudden”
5. L5N5295+1💥	0.83	“ unexpected”“ unfore”“ something”
6. L5N6173+1💥	0.83	“ suddenly”“ myster”“ somehow”
7. L7N8198-1🕒	0.82	“ unexpected”“ situation”“ unexpectedly”
8. L12N5328+1🕒	0.82	“ unexpected”“ discovery”“ randomly”
9. L19N3970-1💤	0.81	“ unexpected”“ surprising”“ unexpectedly”
10. L18N5937-1🕒	0.81	“ unexpectedly”“ suddenly”“ accidentally”

Shadow specimen, apart from the ten: L13N8570+1🕒, 0.21.

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 13, neuron 8570, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.81, top-1 agreement 0.62) 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
L13N8570+1🕒 shadow specimen t0-v0.1	0.878	0.60	1.203	1.37	“ specific”“ particular”“ itself”“ specifics”“ Specific”“ specifically”“ basically”“ directly”
L13N8570-1 t0-v0.1	0.812	0.62	1.031	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.