Layer 6 · Neuron 3428 · boosted

L6N3428+1

Word quirks:

“ funny”“ jokes”“ joke”“ humor”“ hilarious”“ comedy”“ humorous”“ pun”

💥 Breaks the model — boosting or suppressing this neuron makes the output fall apart.

Identity (according to the specimen itself)

I'm sorry, do you want to hear about a monkey?

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. L5N8891-1🕒	0.98	“ humor”“ jokes”“ funny”
2. L4N3483-1💥	0.97	“ funny”“ irony”“ humor”
3. L17N6015-1🕒	0.97	“ jokes”“ humor”“ humorous”
4. L3N5863-1💥	0.96	“ humor”“ witty”“ jokes”
5. L1N4322-1💥	0.96	“ joke”“<\|im_end\|>”“ jokes”
6. L4N5724-1💥	0.95	“ joke”“ jokes”“ joking”
7. L8N2465-1🕒	0.93	“ joking”“ joke”“ amusing”
8. L6N8614+1🕒	0.91	“ comedy”“ comedic”“ comedian”
9. The Class Clown✅ L12N9037-1	0.91	“ humor”“ playful”“ fun”
10. L9N4319+1💤	0.91	“ funny”“ amusing”“ joke”

Shadow specimen, apart from the ten: L6N3428-1💤, 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 6, neuron 3428, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 3.40, top-1 agreement 0.31) 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
L6N3428+1 t0-v0.1	3.400	0.31	3.974	2.52	Top boosted tokens ↑
L6N3428-1💤 shadow specimen t0-v0.1	0.535	0.69	0.643	1.29	“ your”“ now”“ generation”“ my”“ global”“ generated”“ generating”“ central”

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.