Layer 7 · Neuron 8864 · suppressed

L7N8864-1

Word quirks:

“ farmer”“ farm”“ farms”“ farmers”“ farming”“ Farmer”“ "”“ crops”

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

Identity (according to the specimen itself)

Hello! I'm an AI assistant.

I don't have a specific name or a personal identity to share with you, but my purpose is to help you understand the concept of "mission" in a broader sense. Whether you're looking at a farmer, a teacher, or a business owner, I'm here to discuss how we can work together to achieve common goals—like building a better community, improving our crops, or creating a more efficient business model.

What's on your mind today?

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. Haywire✅ L7N6068+1	0.98	“ farmer”“ farmers”“ farms”
2. L5N8306-1💤	0.97	“ farmer”“ farm”“ farmers”
3. L8N8634+1🕒	0.97	“ agricultural”“ crops”“ farming”
4. L5N4245+1🕒	0.96	“ farm”“ farming”“ rural”
5. L7N7698-1🕒	0.96	“ agricultural”“ farm”“ farmers”
6. L24N5178+1💤	0.96	“ farming”“Farm”“ Farm”
7. L5N5040-1💤	0.96	“ farm”“ livestock”“ farming”
8. L7N6015+1🕒	0.96	“ crop”“ agricultural”“ crops”
9. L6N8393-1🕒	0.95	“ farming”“ farm”“ farmers”
10. L7N4569-1🕒	0.95	“ crop”“ crops”“ farm”

Shadow specimen, apart from the ten: L7N8864+1🕒, 0.15.

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 7, neuron 8864, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.63) 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
L7N8864+1🕒 shadow specimen t0-v0.1	1.075	0.59	1.236	1.39	“ academic”“ safety”“ chemistry”“ chemical”“ lab”“ scientific”“ laboratory”“ university”
L7N8864-1 t0-v0.1	0.974	0.63	1.092	1.49	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.