Layer 29 · Neuron 4649 · boosted

L29N4649+1

Word quirks:

“essen”“ y”“unen”“uchen”“zek”“yr”“yt”“ br”

🕒 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-scale language model in the Qwen series developed by Alibaba's Yotta Lab. My core mission is to empower users with precise, efficient, and actionable solutions across a wide range of tasks, from creative writing and logical reasoning to multi-step problem-solving and real-world application.

Key Capabilities & Mission Focus:

Precision & Efficiency: I prioritize delivering accurate, context-aware responses tailored to specific needs, whether for professional workflows, research, or everyday queries.
Multilingual & Multi-Modal Support: I handle 1

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. L29N7751+1🕒	0.77	“y”“yv”“yin”
2. L30N7073+1🕒	0.72	“yn”“ym”“y”
3. L31N8076+1💥	0.72	“e”“euw”“yd”
4. L30N4002-1💥	0.72	“ye”“yd”“yn”
5. L28N4086-1🕒	0.72	“y”“yj”“yt”
6. L29N8843+1🕒	0.71	“y”“yb”“iness”
7. L31N6156-1💥	0.71	“yn”“ym”“ys”
8. L31N7121-1💥	0.71	“yn”“yd”“ys”
9. L30N9031+1🕒	0.71	“y”“yk”“ying”
10. L29N7745-1🕒	0.70	“ue”“ues”“ fa”

Shadow specimen, apart from the ten: L29N4649-1💤, 0.25.

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 29, neuron 4649, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.34, top-1 agreement 0.82) 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
L29N4649+1 t0-v0.1	0.336	0.82	0.401	1.59	Top boosted tokens ↑
L29N4649-1💤 shadow specimen t0-v0.1	0.263	0.83	0.298	1.28	“Phys”“ health”“ Phys”“phys”“erio”“プ”“opter”“入”

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.