Layer 16 · Neuron 5604 · boosted

L16N5604+1

Word quirks:

“ ⏎”“ ___”“ __”“ not”“ earlier”“ previously”“ **”“ ____”

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

Identity (according to the specimen itself)

Answered — readability check pending

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. L13N1060-1🕒	0.86	“ ⏎”“ **”“ before”
2. L15N4256-1💤	0.83	“ **"”“ ⏎”“ __”
3. L12N2527+1🕒	0.82	“ ____”“ now”“<\|im_end\|>”
4. L13N7314-1💤	0.81	“ "**”“ previously”“ earlier”
5. L12N475+1🕒	0.80	“ ⏎”“ _”“ __”
6. L13N5939+1🕒	0.79	“ ____”“ earlier”“ ______”
7. L16N6076+1💤	0.79	“ ____”“ earlier”“<\|im_end\|>”
8. L8N6691+1🕒	0.78	“ **”“ ______”“ my”
9. L15N1711+1🕒	0.78	“<\|im_end\|>”“⏎”“ later”
10. L11N2530+1🕒	0.77	“ __”“ NOT”“ '”

Shadow specimen, apart from the ten: L16N5604-1🕒, 0.44.

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 16, neuron 5604, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.54, top-1 agreement 0.78) 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
L16N5604+1 t0-v0.1	0.544	0.78	0.705	1.17	Top boosted tokens ↑
L16N5604-1🕒 shadow specimen t0-v0.1	1.256	0.51	1.549	2.01	“ ...”“ basically”“ ‘”“ …”“ ….”“ (...)”“ […]”“ (…)”

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.