Layer 30 · Neuron 8695 · suppressed

L30N8695-1

Word quirks:

“icot”“耽美”“夸”“礼物”“ теч”“kör”“impa”“ériences”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

I am Qwen3.5, the latest large language model developed by Tongyi Lab. My primary mission is to empower users with advanced, practical, and reliable AI assistance across a wide spectrum of tasks. Here's how I aim to deliver value:

Core Capabilities & Mission Focus:

Advanced Reasoning & Problem-Solving

Tackle complex math, science, and logical challenges with step-by-step precision.
Analyze multi-step workflows for tasks like debugging code or financial planning.

Deep Expertise in Specialized Domains

Provide accurate,

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. L20N8601+1💤	0.68	“电器”“avour”“萍”
2. L26N3205+1💤	0.66	“吃”“ogat”“靶”
3. L24N5049+1💤	0.65	“edel”“奉行”“ORO”
4. L25N7594-1🕒	0.65	“酷”“ pomp”“包”
5. L30N2093+1💤	0.65	“impat”“空气能”“imitives”
6. L28N6506+1💤	0.65	“过节”“amour”“gam”
7. L24N381-1💤	0.65	“器材”“第二”“教主”
8. L28N875-1💤	0.64	“畜”“抬”“othi”
9. L29N5129+1🕒	0.64	“拖拉”“娇”“.ke”
10. L30N3956-1💤	0.64	“賞”“ambang”“ Lago”

Shadow specimen, apart from the ten: L30N8695+1💥, 0.09.

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 30, neuron 8695, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.31, 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
L30N8695+1💥 shadow specimen t0-v0.1	5.914	0.04	5.989	1.02	“ indices”“指标”“ Indices”“ indexed”“ indexes”“ Jones”“indices”“Indices”
L30N8695-1 t0-v0.1	0.310	0.82	0.339	1.66	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.