Layer 10 · Neuron 3397 · boosted

L10N3397+1

Word quirks:

“ confusion”“ despair”“ frustration”“ sadness”“ misery”“ chaos”“ desperation”“ paranoia”

🕒 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 am Qwen3. I am designed to be helpful.

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. L11N5393-1🕒	0.89	“ despair”“ melanch”“ sorrow”
2. The Meltdown✅ L5N5774+1	0.84	“ exhaustion”“ tears”“ agony”
3. L4N4651+1🕒	0.84	“ sadness”“ pain”“ punishment”
4. L2N7385-1💥	0.81	“ misery”“ miserable”“<\|im_end\|>”
5. L4N5229+1💥	0.81	“ sorrow”“ lament”“ anguish”
6. L11N8992+1🕒	0.81	“ desperate”“ desperately”“ exhausted”
7. L6N8474-1💥	0.80	“ madness”“ Kafka”“ existence”
8. L2N3829-1💥	0.79	“ bleak”“ despair”“ miserable”
9. L4N8515-1💥	0.79	“ sorrow”“ anguish”“ pain”
10. L1N7617-1💥	0.78	“ distress”“ distressed”“<\|im_end\|>”

Shadow specimen, apart from the ten: L10N3397-1💤, -0.07.

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 10, neuron 3397, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.95, top-1 agreement 0.40) 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
L10N3397+1 t0-v0.1	1.954	0.40	2.247	2.22	Top boosted tokens ↑
L10N3397-1💤 shadow specimen t0-v0.1	0.743	0.68	0.971	1.40	“ ‘”“ “”“ better”“ **“”“ more”“ my”“ referring”“ myself”

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.