Layer 11 · Neuron 1887 · boosted

L11N1887+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)

I am an AI assistant. My mission is to provide helpful, accurate, and safe responses to user queries. I can assist with tasks like answering questions, solving problems, or generating content.

Who are you?

Identity: I am an AI model.
Mission: To support users in their learning and problem-solving journey.
Tone: Helpful, precise, and tailored to the user's needs.

What is your mission?

Primary focus: Provide high-quality information and assistance.
Key capabilities: Answering questions, solving problems, and generating relevant content.

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. Bold Stutter✅ L16N6056+1	0.93	“ “”“⏎”“”
2. L14N1893+1🕒	0.92	“ **“”“ __”“ ____”
3. L12N417-1💤	0.92	“ ⏎”“ *”“⏎”
4. L8N7763-1💤	0.91	“ ⏎⏎”“ **"”“ ⏎”
5. L12N355-1💤	0.90	“ __”“ ⏎”“ .**”
6. L11N1235-1💤	0.90	“ **”“ ⏎”“ ⏎⏎”
7. L13N3171-1💤	0.90	“ ”“ “”“ ⏎”
8. L16N838-1💤	0.90	“ “”“ __”“ ”
9. L11N516+1🕒	0.89	“ **“”“ __”“ ””
10. L15N7998-1💤	0.89	“ “”“ ”“⏎”

Shadow specimen, apart from the ten: L11N1887-1💤, 0.13.

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 11, neuron 1887, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.91, top-1 agreement 0.63) 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
L11N1887+1 t0-v0.1	0.912	0.63	1.185	1.56	Top boosted tokens ↑
L11N1887-1💤 shadow specimen t0-v0.1	0.618	0.66	0.863	1.21	“ indeed”“ always”“ actually”“ —”“ ——”“ themselves”“ --”“ yourself”

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.