Layer 6 · Neuron 4813 · suppressed

L6N4813-1

Word quirks:

“⏎”“ __”“ **“”“ ______”“ ‘”“ ____”“⏎ ⏎”“ *“”

💤 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 in the Qwen series. I am designed to be your most capable assistant, offering enhanced capabilities across a wide range of domains.

🌟 My Mission:

To provide accurate, reliable, and insightful support for users by leveraging my advanced knowledge base, reasoning abilities, and multimodal tools. Whether you need help with complex tasks, creative problem-solving, or deep analysis, I strive to deliver high-quality results tailored to your needs.

---

🔍 Key Capabilities:

Advanced Reasoning:

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. Trailer Voice✅ L5N2333+1	0.96	“ *“”“ ‘”“ “”
2. L8N1847-1🕒	0.96	“⏎”“ **“”“ ______”
3. L4N1109+1🕒	0.95	“⏎”“ ‘”“ ____”
4. L6N6139-1🕒	0.95	“⏎”“ **“”“ ‘”
5. L3N1947+1💤	0.95	“⏎”“ ‘”“ ____”
6. L6N5823-1💤	0.95	“⏎”“ ‘”“ **“”
7. L2N6255-1💤	0.95	“ ‘”“⏎”“ “”
8. L3N2857-1🕒	0.95	“ “”“ ‘”“⏎”
9. L4N1236+1💤	0.95	“⏎”“ ‘”“ “”
10. L13N2353+1💤	0.95	“ **“”“ ____”“ ______”

Shadow specimen, apart from the ten: L6N4813+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 6, neuron 4813, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.58, top-1 agreement 0.67) 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
L6N4813+1🕒 shadow specimen t0-v0.1	1.390	0.52	1.712	1.57	“⏎”“<\|im_end\|>”“\n”“ info”“ ____”“⏎⏎⏎”“...\”“!”
L6N4813-1 t0-v0.1	0.582	0.67	0.693	1.37	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.