Layer 4 · Neuron 9135 · suppressed

L4N9135-1

Word quirks:

“ thinking”“ words”“ phrase”“ “”“ name”“ ‘”“ thoughts”“!”

💤 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 assistant developed by Google DeepMind. My mission is to assist users with answering questions, solving problems, or generating ideas across a wide range of topics. I strive to provide helpful, accurate, and insightful responses to user inquiries.

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. L5N8177-1🕒	0.81	“ “”“ ‘”“ thinking”
2. L4N7129+1🕒	0.79	“ thinking”“ “”“ metaphor”
3. L5N5811+1💤	0.79	“ ‘”“ —”“ thinking”
4. L4N6814+1🕒	0.78	“ ‘”“ “”“ words”
5. L4N8375-1🕒	0.75	“ ‘”“ “”“ thinking”
6. L3N6078-1🕒	0.75	“ “”“ ””“ ‘”
7. L4N8282-1🕒	0.74	“ name”“ “”“ words”
8. L4N4347-1🕒	0.74	“ reason”“ thinking”“ ‘”
9. L4N7870+1🕒	0.74	“ thinking”“ ‘”“ thought”
10. L17N4816+1💤	0.73	“ “”““”“ thinking”

Shadow specimen, apart from the ten: L4N9135+1🕒, 0.22.

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 4, neuron 9135, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.84, top-1 agreement 0.68) 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
L4N9135+1🕒 shadow specimen t0-v0.1	2.157	0.49	2.560	2.36	“ impacts”“ impact”“ experience”“ change”“ ‘”“ experiences”“ changes”“ affect”
L4N9135-1 t0-v0.1	0.841	0.68	1.031	1.46	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.