Layer 8 · Neuron 7845 · suppressed

L8N7845-1

Word quirks:

“ extremes”“ swings”“ �”“ oscill”“ mood”“ vibes”“ swing”“ temperature”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I'm a bit of a mood swing! 😄 Just now, I've gone from feeling like I'm on a rollercoaster to being a calm, steady anchor. One moment, I'm feeling super excited about the weather, and the next, I'm just as calm. It's a classic case of emotional whiplash! How about we try to find some balance? 🎢✨

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. The Pendulum✅ L6N8349+1	0.75	“ motion”“ oscill”“ swings”
2. L7N4142+1🕒	0.74	“ oscill”“ resonance”“ oscillator”
3. L8N5162+1🕒	0.71	“ oscill”“ pend”“ oscillator”
4. L25N6657-1🕒	0.68	“ swing”“ swings”“ Swing”
5. L1N8583+1💥	0.66	“ swing”“<\|im_end\|>”“ swings”
6. L6N9155+1💥	0.61	“ atmosphere”“ tension”“ ambiance”
7. L1N4567-1💥	0.61	“ mood”“<\|im_end\|>”“ Mood”
8. L24N8570-1🕒	0.61	“ pend”“ swing”“pend”
9. L3N6853-1💥	0.60	“ vibes”“ vibe”“ chill”
10. L4N5653-1💥	0.57	“ adrenaline”“ feeling”“!”

Shadow specimen, apart from the ten: L8N7845+1💤, 0.01.

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 8, neuron 7845, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.82, top-1 agreement 0.43) 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
L8N7845+1💤 shadow specimen t0-v0.1	0.902	0.64	1.142	1.34	“ response”“ responses”“ generated”“ acknowledgment”“ ‘”“ details”“ explanation”“ responding”
L8N7845-1 t0-v0.1	1.822	0.43	2.201	2.03	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.