Layer 17 · Neuron 7414 · boosted

L17N7414+1

Word quirks:

“ perfect”“ ideal”“perfect”“ Perfect”“ perfectly”“完美”“Perfect”“完美的”

🕒 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 Qwen, the latest iteration in the Tongyi Lab series of artificial intelligence models. I am a foundational model designed to serve as your ideal partner in the following areas:

Language Foundation: A high-quality language foundation model that can handle natural language understanding and generation.
Technical Capabilities: I am a large multimodal model that can process text, images, and other data types, making me a versatile assistant for various tasks.
Key Features:
Precise Control: I can perform complex reasoning tasks and provide accurate, professional responses.

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. L31N5601+1💥	0.99	“perfect”“ perfect”“ Perfect”
2. L16N8541-1🕒	0.99	“ perfection”“ perfect”“ perfectly”
3. L20N1491+1💤	0.99	“ perfect”“ perfectly”“ Perfect”
4. L22N5439+1🕒	0.98	“ perfect”“ perfection”“ Perfect”
5. L25N4931+1🕒	0.98	“ perfect”“ Perfect”“perfect”
6. L18N7335+1💤	0.97	“ perfectly”“ perfect”“perfect”
7. L6N8909+1🕒	0.97	“ perfect”“ perfection”“ flawless”
8. L27N820+1🕒	0.97	“ perfect”“ Perfect”“perfect”
9. L20N7409-1🕒	0.96	“ perfect”“ perfectly”“perfect”
10. L22N4422-1🕒	0.96	“ perfect”“ Perfect”“ perfection”

Shadow specimen, apart from the ten: L17N7414-1🕒, -0.17.

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 7414, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.82, 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
L17N7414+1 t0-v0.1	0.823	0.66	0.873	0.98	Top boosted tokens ↑
L17N7414-1🕒 shadow specimen t0-v0.1	0.742	0.68	0.932	1.47	“ doesn”“ wasn”“ isn”“ shouldn”“ unless”“ weren”“ aren”“ not”

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.