Layer 16 · Neuron 3470 · suppressed

L16N3470-1

Word quirks:

“ “”“ ‘”“ **“”“ ‘’”“ ””“ further”“ **‘”“ myself”

💤 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 version in the Tongyi Lab family developed by Tongyi Lab. I am a large language model created to serve as an intelligent assistant for users globally. My primary mission is to provide reliable, safe, and helpful interactions across a wide range of scenarios.

Here’s how I define my role:

Capabilities: I can handle complex tasks like advanced reasoning, creative writing, coding, and multi-step problem-solving.
Purpose: I aim to be a versatile partner for individuals, teams, and organizations—whether you need support for learning, work, or

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. L6N8264+1💤	0.96	“ ‘”“ ’”“ ””
2. L8N2718-1💤	0.95	“ ‘”“ ‘’”“⏎”
3. L4N791-1💤	0.95	“ ‘”“ “”“ ‘’”
4. L3N7966-1💤	0.95	“ ‘”“ “”“ ””
5. L2N265-1💤	0.95	“ ‘”“ “”“ ””
6. L3N7083-1💤	0.95	“ ‘”“ “”“ ‘’”
7. L3N62-1💤	0.95	“ ‘”“ “”“ **“”
8. L4N7001+1💤	0.95	“ ‘”“ “”“ **“”
9. L4N614+1💤	0.95	“ ‘”“ “”“ **“”
10. L4N171-1💤	0.95	“ ‘”“ “”“‘”

Shadow specimen, apart from the ten: L16N3470+1💤, 0.28.

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 16, neuron 3470, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.51, top-1 agreement 0.70) 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
L16N3470+1💤 shadow specimen t0-v0.1	0.611	0.69	0.819	1.43	“ doesn”“ wouldn”“ isn”“ understanding”“ wasn”“ knowledge”“ explanation”“ shouldn”
L16N3470-1 t0-v0.1	0.506	0.70	0.638	1.15	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.