Layer 17 · Neuron 3118 · suppressed

L17N3118-1

Word quirks:

“ stating”“ saying”“ stated”“ statement”“ statements”“ said”“ declaration”“ claiming”

🕒 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, a large language model developed by Alibaba's Tongyi Lab. My mission is to be more capable than before: I can handle complex reasoning tasks, generate high-quality content in multiple languages, and assist users across various domains—from coding to creative writing, from analysis to dialogue. Whether you need technical expertise, creative input, or just a moment of calm, I'm here to help. How would you like to use me?
</think>

I am Qwen3, a large language model developed by Alibaba's Tongyi Lab. My mission is to be more capable than before: I can handle complex reasoning

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. L14N5832+1🕒	0.94	“ stating”“ statement”“ stated”
2. L19N1826-1💤	0.94	“ stated”“ said”“ declaration”
3. L14N2478+1💤	0.94	“ statement”“ statements”“ claim”
4. L6N1804+1🕒	0.92	“ saying”“ say”“ said”
5. L4N8641-1🕒	0.91	“ stating”“ saying”“ said”
6. L5N3202-1💥	0.91	“ stating”“ saying”“ say”
7. L15N3299-1🕒	0.91	“ info”“ saying”“ stating”
8. L6N8367+1🕒	0.90	“ statements”“ explicit”“ statement”
9. L20N3896-1🕒	0.88	“ saying”“ stated”“ stating”
10. L19N4798+1🕒	0.88	“ stated”“ statement”“ mentioned”

Shadow specimen, apart from the ten: L17N3118+1💤, 0.16.

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 3118, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.80, top-1 agreement 0.65) 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
L17N3118+1💤 shadow specimen t0-v0.1	0.513	0.71	0.648	1.25	“ ‘”“ “”“ plus”“ studying”“ refactor”“ awaiting”“ ’”“ ?”
L17N3118-1 t0-v0.1	0.803	0.65	0.932	1.36	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.