Layer 12 · Neuron 1684 · suppressed

L12N1684-1

Word quirks:

“⏎”“⏎⏎”“ ?”“?\”“?”“ ⏎”“⏎ ⏎”“?"”

🕒 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 a large language model trained by Google. The company's mission is to provide high-quality AI tools that help everyone achieve more, and empower every person to better express themselves.

If you have a specific task, feel free to ask!

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. L16N1298+1💤	0.96	“⏎”“<\|im_end\|>”“⏎⏎”
2. L8N2618+1🕒	0.95	“⏎”“<\|im_end\|>”“⏎⏎”
3. L3N3824+1💤	0.95	“⏎”“<\|im_end\|>”“⏎⏎”
4. L4N1936+1🕒	0.95	“<\|im_end\|>”“⏎”“?\”
5. L5N585-1🕒	0.95	“⏎”“<\|im_end\|>”“?\”
6. L1N2041-1💥	0.95	“<\|im_end\|>”“⏎”“⏎ ⏎”
7. L5N8688+1🕒	0.95	“<\|im_end\|>”“?\”“⏎”
8. L3N377+1🕒	0.95	“<\|im_end\|>”“⏎”“⏎⏎”
9. L3N335+1🕒	0.95	“<\|im_end\|>”“⏎”“?\”
10. The Waffler✅ L12N6462+1	0.95	“?”“?\”“⏎”

Shadow specimen, apart from the ten: L12N1684+1💤, -0.19.

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 12, neuron 1684, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.87, top-1 agreement 0.65) 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
L12N1684+1💤 shadow specimen t0-v0.1	0.729	0.69	0.975	1.31	“ sometimes”“ similar”“ anyway”“ accordingly”“ always”“ now”“ even”“ similarly”
L12N1684-1 t0-v0.1	0.867	0.65	1.070	1.34	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.