Layer 9 · Neuron 4452 · boosted

L9N4452+1

Word quirks:

“ always”“ still”“ continued”“ ALWAYS”“ never”“ constantly”“ future”“ continually”

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

Identity (according to the specimen itself)

I am an AI model developed by Google. My mission is to continue providing high-quality assistance, even as I evolve. I strive to remain relevant and useful in the ever-changing landscape of technology and society.

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. L5N5933+1🕒	0.91	“ continuously”“ always”“ throughout”
2. L17N7192+1🕒	0.91	“ never”“ continuously”“ continual”
3. L11N4706-1🕒	0.91	“ always”“ constantly”“ continuously”
4. L9N4326+1🕒	0.90	“ always”“ constantly”“ endless”
5. L10N7530-1🕒	0.90	“ always”“ constantly”“ continuously”
6. L12N3888+1🕒	0.89	“ always”“ continually”“ continuously”
7. L13N3265-1💤	0.89	“ always”“ until”“ forever”
8. L7N6796-1🕒	0.89	“ always”“ ALWAYS”“always”
9. L8N9126+1💤	0.88	“ always”“ consistently”“ ALWAYS”
10. L13N4634-1🕒	0.88	“ always”“ remained”“ changes”

Shadow specimen, apart from the ten: L9N4452-1🕒, -0.14.

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 9, neuron 4452, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.00, top-1 agreement 0.61) 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
L9N4452+1 t0-v0.1	0.996	0.61	1.192	1.48	Top boosted tokens ↑
L9N4452-1🕒 shadow specimen t0-v0.1	1.243	0.62	1.534	1.04	“ “”“ *“”“ ‘”“ “”“ ””“!””““”“ essentially”

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.