Layer 17 · Neuron 2256 · suppressed

L17N2256-1

Word quirks:

“$(”“ $(”“${”“ typical”“ ${”“$”“⏎⏎”“””

🕒 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 Qwen1.5, a large language model designed to enhance human-AI interaction by providing high-quality responses across various domains. My primary mission is to assist users in completing tasks, generate text, and facilitate learning or problem-solving. I aim to be a versatile and efficient tool that adapts to different contexts, offering accurate information, creative writing, coding assistance, and logical reasoning capabilities. By leveraging my training data and architectural improvements, I strive to maximize performance and contribute to a more productive and engaging user experience.

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. L31N6969-1💥	0.86	“${”“$(”“ ${”
2. L31N615+1💥	0.86	“ $(”“ $\”“ ${”
3. L24N1047-1💤	0.85	“$(”“ $(”“ (${”
4. L21N327-1💤	0.84	“ $(”“ $\”“$\”
5. L21N1388-1💤	0.84	“ $(”“$(”“ $($”
6. L19N2248+1💤	0.83	“${”“$”“{”
7. L6N43+1🕒	0.81	“⏎ ⏎”“ ⏎ ⏎”“$”
8. L24N5790-1🕒	0.80	“$(”“ $(”“ "$(”
9. L31N6988+1💥	0.79	“$(”“ $(”“](”
10. L21N722-1💤	0.78	“ $”“ `$”“ $(”

Shadow specimen, apart from the ten: L17N2256+1💤, 0.01.

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 2256, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.66, top-1 agreement 0.67) 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
L17N2256+1💤 shadow specimen t0-v0.1	0.511	0.72	0.585	1.18	“ Global”“ Globe”“ global”“ Glob”“ globally”“ Worldwide”“ Coverage”“ Universal”
L17N2256-1 t0-v0.1	0.663	0.67	0.798	1.22	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.