Layer 17 · Neuron 5767 · boosted

L17N5767+1

Word quirks:

“ stated”“ specified”“ mentioned”“ said”“ described”“ stating”“ fulfilling”“ that”

🕒 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 version specified by the prompt. I am a large language model with capabilities in reasoning, coding, and knowledge base, designed to assist users effectively. My mission is to provide high-quality support by enhancing problem-solving, creativity, and technical proficiency while adhering to safety standards. How can I assist you?

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. L23N2571-1💤	0.90	“ said”“ stated”“ mentioned”
2. L16N5036-1🕒	0.88	“ stated”“ provided”“ said”
3. L16N3501-1💤	0.88	“ mentioned”“ referenced”“ stated”
4. L16N1670-1💤	0.88	“ said”“ stated”“ suggested”
5. L12N1317+1🕒	0.87	“ *”“ said”“ mentioned”
6. L12N1493-1🕒	0.86	“ specified”“ described”“ referenced”
7. L16N9031-1🕒	0.85	“ stated”“ described”“ noted”
8. L16N7011+1🕒	0.85	“ mentioned”“ stated”“ said”
9. L16N7694+1🕒	0.85	“ mentioned”“ *”“ said”
10. L19N4798+1🕒	0.85	“ stated”“ statement”“ mentioned”

Shadow specimen, apart from the ten: L17N5767-1💤, 0.18.

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 5767, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, 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
L17N5767+1 t0-v0.1	0.760	0.66	0.889	1.27	Top boosted tokens ↑
L17N5767-1💤 shadow specimen t0-v0.1	0.435	0.75	0.520	0.97	“ called”“ call”“ known”“ also”“ calls”“called”“ Called”“叫什么”

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.