Layer 29 · Neuron 7988 · boosted

L29N7988+1

Word quirks:

“ vi”“ Vi”“vi”“喋”“ính”“angu”“ényt”“恶”

🕒 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, a large language model developed by Tongyi Lab. My primary mission is to be a versatile and reliable AI assistant that empowers users across diverse domains:

Key Capabilities & Focus Areas:

Deeply Human-Centric: I prioritize natural, context-aware, and ethical interactions, adapting my tone and style to different users and scenarios.
Reasoning & Problem-Solving: I excel in complex tasks like mathematical computation, scientific analysis, and multi-step logical reasoning.
Long Context Mastery: I can process and deeply understand extremely long documents (e

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. L29N8536+1🕒	0.74	“ vi”“ Vi”“vi”
2. L29N6900-1🕒	0.73	“vi”“ vi”“ vy”
3. The Salesman✅ L21N748-1	0.72	“ Vi”“kur”“vi”
4. L30N1602+1💥	0.72	“ vi”“ Vi”“vi”
5. L26N7332+1🕒	0.71	“ Vi”“Vi”“维”
6. L30N1733-1💤	0.71	“Vi”“�”“ Vi”
7. L27N5950-1🕒	0.68	“ vi”“ Vi”“Vi”
8. L28N2773-1💤	0.67	“ venn”“ vi”“智”
9. L27N3924-1💤	0.67	“ VI”“VI”“ Vi”
10. L28N9127+1🕒	0.67	“ vi”“vi”“ val”

Shadow specimen, apart from the ten: L29N7988-1🕒, 0.15.

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 7988, 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 0.77) 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
L29N7988+1 t0-v0.1	0.335	0.77	0.402	1.38	Top boosted tokens ↑
L29N7988-1🕒 shadow specimen t0-v0.1	0.384	0.80	0.452	1.64	“eld”“ biz”“elden”“cel”“leve”“ammo”“uang”“aires”

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.