Layer 13 · Neuron 1167 · boosted

L13N1167+1

Word quirks:

“ typical”“ usually”“ usual”“ mentioned”“ here”“ referenced”“ you”“ typically”

🕒 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 high-quality support across various domains:

Technical Solutions: I can help with technical documentation, code snippets, and system architecture discussions.
Knowledge Sharing: For general inquiries, I'll share relevant information in a structured format.
Collaboration: When working on projects together, I'll ensure all deliverables meet our agreed-upon standards through clear communication channels.

Please let me know if you'd like to discuss any specific topics or tasks. I'm here to assist with your work-related needs! 😊

Note: This is

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. L9N2209-1💤	0.89	“ usually”“ specific”“ this”
2. L9N6651+1🕒	0.88	“ your”“ ideal”“ typically”
3. L14N5463-1💤	0.88	“ typically”“ typical”“ usual”
4. L15N2730-1🕒	0.88	“ typically”“ usually”“ usual”
5. L9N3385+1🕒	0.88	“ commonly”“ typically”“ known”
6. L16N465-1💤	0.88	“ commonly”“ often”“ typical”
7. L11N4813-1💤	0.87	“ usually”“ sometimes”“ you”
8. L12N4598+1💤	0.86	“ typically”“ typical”“ usually”
9. L10N3796-1💤	0.86	“ typically”“ usually”“ commonly”
10. L13N7813-1🕒	0.86	“ usually”“ often”“ typically”

Shadow specimen, apart from the ten: L13N1167-1🕒, -0.12.

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 13, neuron 1167, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.97, top-1 agreement 0.60) 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
L13N1167+1 t0-v0.1	0.973	0.60	1.203	1.54	Top boosted tokens ↑
L13N1167-1🕒 shadow specimen t0-v0.1	1.360	0.58	1.704	1.40	“<\|im_end\|>”“:\”“ \"”“ ⏎”“�”“):\”“:\"”“.\”

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.