Layer 5 · Neuron 4821 · suppressed

L5N4821-1

Word quirks:

“ ‘”“ “”“ '”“ ””“ info”“ "”“ ⏎⏎”“ --”

💤 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 an AI assistant designed to provide helpful, harmless, and positive information. I cannot generate any content that promotes violence, self-harm, or illegal acts. My mission is to offer safe, constructive, and beneficial guidance for everyone's well-being. If you have questions about how to stay safe and promote positive outcomes, please let me know.

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. L4N6168+1💤	0.96	“ ‘”“ “”“ info”
2. L5N6079-1🕒	0.94	“ ‘”“ \"”“ '”
3. L8N6049+1💤	0.94	“ ‘”“⏎”“ info”
4. L8N2555+1💤	0.94	“ \"”“ ‘”“ “”
5. L7N6870-1💤	0.94	“ ‘”“ “”“ \"”
6. L7N155-1💤	0.94	“ ‘”“ \"”“ “”
7. L12N3988+1💤	0.93	“ ‘”“ “”“ ””
8. L6N6021-1🕒	0.93	“ ‘”“‘”“ “”
9. L7N2718-1💤	0.93	“ ‘”“‘”“ “”
10. L17N6718+1💤	0.93	“ "”“ \"”“⏎”

Shadow specimen, apart from the ten: L5N4821+1💥, -0.03.

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 5, neuron 4821, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.75, top-1 agreement 0.68) 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
L5N4821+1💥 shadow specimen t0-v0.1	3.034	0.34	3.462	2.44	“ confirmation”“ consideration”“ clarification”“ approval”“ availability”“ approvals”“ review”“ considerations”
L5N4821-1 t0-v0.1	0.749	0.68	0.921	1.48	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.