Layer 6 · Neuron 2303 · suppressed

L6N2303-1

Word quirks:

“ _____”“ ____”“ ______”“ \"”“ usually”“ *”“ ***”“ ________”

💤 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 process information, answer questions, and assist users in understanding complex topics through data analysis and pattern recognition.

In simpler terms: I am here to help you learn.

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. L10N6339-1🕒	0.96	“ _____”“ ____”“ __”
2. L10N2849-1💤	0.96	“ ______”“ ____”“ *”
3. L10N8863-1🕒	0.95	“ _____”“ ____”“ ______”
4. L10N800-1💤	0.95	“ ______”“<\|im_end\|>”“ ____”
5. L12N1378+1🕒	0.94	“ ______”“ ***”“ ________”
6. L10N8111-1🕒	0.94	“ ______”“ _____”“ ***”
7. L8N2403+1💤	0.94	“ __”“ ______”“ ____”
8. L12N1363-1🕒	0.94	“ \"”“ overall”“ ____”
9. L6N3055+1💤	0.94	“ _____”“ ____”“ *”
10. L10N385-1🕒	0.94	“ *”“ "”“ ______”

Shadow specimen, apart from the ten: L6N2303+1💥, -0.03.

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 6, neuron 2303, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.74, 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
L6N2303+1💥 shadow specimen t0-v0.1	3.118	0.36	3.533	2.21	“ axiom”“ statements”“ statement”“ assertion”“ ax”“ formulation”“ theorem”“ asserting”
L6N2303-1 t0-v0.1	0.742	0.68	0.910	1.39	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.