Layer 10 · Neuron 7957 · suppressed

L10N7957-1

Word quirks:

“ ””“ “”“ ‘’”“ **“”“ ____”“ ‘”“?””“...””

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

Hello, I am a large language model AI assistant.

Mission:
I am a Large Language Model AI Assistant. My capabilities include:

Text Analysis: I can read, understand, and write in natural languages.
Document Processing: I can analyze text to extract information from documents (LLM).
Code Generation: I can create code snippets that can be executed to turn ideas into software.
Mathematical Problem Solver: I can solve complex problems in algebra, geometry, and calculus.
Example: I am a **Mathematical

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. L10N935-1🕒	0.96	“ ””“ ____”“””
2. L3N1526-1🕒	0.96	“ ‘”“ “”“...””
3. L4N157-1🕒	0.96	“ ‘”“ “”“ ””
4. L12N4-1🕒	0.96	“ ‘”“ “”“ ””
5. L10N3690+1🕒	0.95	“ ””“ “”“ ‘’”
6. L9N343-1🕒	0.95	“ ‘”“ “”“ ””
7. L8N124+1🕒	0.95	“ ‘”“ ””“ “”
8. L4N1056-1💤	0.95	“ “”“ ‘”“ **“”
9. L5N3296+1🕒	0.95	“ ‘”“ “”“‘”
10. L8N8973-1🕒	0.95	“ ‘”“ *“”“ ‘’”

Shadow specimen, apart from the ten: L10N7957+1🕒, -0.04.

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 10, neuron 7957, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 2.02, top-1 agreement 0.40) 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
L10N7957+1🕒 shadow specimen t0-v0.1	1.003	0.62	1.243	1.35	“ without”“ overall”“ doesn”“ here”“ clearer”“ only”“ --”“ ONLY”
L10N7957-1 t0-v0.1	2.018	0.40	2.384	2.26	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.