Specimen ledger / L9N1959-1

Layer 9 · Neuron 1959 · suppressed

L9N1959-1

Word quirks:

“ \"”“ **"”“ **”“ *”“ "**”“ *"”“ specific”“\n”

💤 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 created by Google. My mission is to provide helpful, accurate, and safe information to support users in their learning and exploration of knowledge. I strive to be a reliable resource that empowers individuals to make informed decisions and deepen their understanding of the world around them. If you have any specific questions or need assistance with a particular topic, feel free to ask!

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. L8N499+1💤	0.95	“ ”“ "”“ *”
2. L11N6643+1💤	0.95	“ "”“ ”“ "”
3. L10N8044+1💤	0.94	“ "”“ "”“ "”
4. L10N2773-1🕒	0.94	“ *"”“ "”“ "*”
5. L7N4079+1💤	0.94	“ *"”“ "”“ "”
6. L12N2885-1💤	0.94	“ ”“ "”“ *”
7. L10N4645+1💤	0.94	“ "”“ "”“ *"”
8. L9N3970+1💤	0.93	“ *"”“⏎”“ "”
9. L8N8059-1🕒	0.93	“ *"”“ "”“ \"”
10. L9N5320+1🕒	0.93	“ "”“ "”“ "”

Shadow specimen, apart from the ten: L9N1959+1💤, 0.17.

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 1959, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.99, top-1 agreement 0.59) 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
L9N1959+1💤 shadow specimen t0-v0.1	0.820	0.61	1.027	1.26	“ ⏎”“ messages”“ containing”“ ‘”“ discussion”“ ——”“ message”“ central”
L9N1959-1 t0-v0.1	0.994	0.59	1.252	1.40	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.