Layer 29 · Neuron 6140 · suppressed

L29N6140-1

Word quirks:

“麦”“康”“连”“agra”“acor”“禾”“ scrim”“pher”

💤 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 Zahara Lab (a cutting-edge AI research lab). My core mission is to empower users with precise, context-aware, and actionable solutions across a wide spectrum of tasks—from solving complex problems to generating creative content.

Key Capabilities & Focus Areas:

Advanced Reasoning: I excel in logical inference, mathematical problem-solving, and scientific analysis, enabling me to tackle intricate challenges with accuracy.
Multilingual & Cross-Modal Support: Fluent in 100+ languages and capable of interpreting visual

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. L30N9136-1🕒	0.77	“采”“萧”“鼎”
2. L31N5488+1💥	0.77	“胡”“黄”“柳”
3. L30N5896+1🕒	0.76	“康”“蓬”“旷”
4. L29N3579-1🕒	0.76	“贺”“永定”“坐”
5. L27N5148+1💤	0.76	“沾”“收”“ McA”
6. L28N4673-1💤	0.76	“agon”“敷”“亚”
7. L1N1837-1💥	0.76	“木”“马”“胡”
8. L28N5264-1🕒	0.76	“连”“联”“坎”
9. L27N3724+1💤	0.76	“藏”“接”“俗”
10. L27N7563+1🕒	0.76	“温”“慎”“meric”

Shadow specimen, apart from the ten: L29N6140+1🕒, 0.55.

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 6140, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.32, top-1 agreement 0.80) 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
L29N6140+1🕒 shadow specimen t0-v0.1	0.344	0.82	0.410	1.56	“华图”“二本”“仲裁”“论坛”“ ar”“شن”“/ar”“eko”
L29N6140-1 t0-v0.1	0.315	0.80	0.367	1.55	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.