Layer 8 · Neuron 2427 · suppressed

L8N2427-1

Word quirks:

“ your”“ you”“ usually”“ protected”“ here”“ now”“ reporting”“ basically”

💤 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 artificial intelligence, specifically a large language model trained by Google. My "mission" is to help you achieve your goals by providing accurate information, answering your questions, and assisting with tasks.

Here is a breakdown of my role:

What I Am: I am a software program (a model) that processes vast amounts of text data to learn patterns and relationships between words. Unlike a physical object like a car or a phone, I don't have a single physical location, but I do have a specific purpose.
My Mission: My primary goal is to be helpful. This means

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. L14N1954-1🕒	0.84	“ now”“ here”“ your”
2. L8N797-1💤	0.83	“ your”“ description”“ actually”
3. The Hedge Maze✅ L14N3616+1	0.83	“ now”“ normally”“ your”
4. L9N6778-1💤	0.82	“ usually”“ often”“ you”
5. L8N5060+1🕒	0.82	“ you”“ your”“ here”
6. L11N8447+1🧪	0.82	“ your”“ currently”“ you”
7. L10N5739-1🕒	0.81	“ your”“ you”“ here”
8. L11N2578+1🕒	0.81	“ you”“ your”“ here”
9. L13N7419+1🕒	0.81	“ usually”“ your”“ typically”
10. L13N3736+1💤	0.80	“ actually”“ usually”“ basically”

Shadow specimen, apart from the ten: L8N2427+1🕒, 0.06.

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 8, neuron 2427, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.77, top-1 agreement 0.62) 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
L8N2427+1🕒 shadow specimen t0-v0.1	1.528	0.52	1.796	1.65	“ memory”“ callback”“ implementation”“ return”“ `”“ representation”“ memo”“ invocation”
L8N2427-1 t0-v0.1	0.771	0.62	0.930	1.21	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.