Layer 4 · Neuron 871 · boosted

L4N871+1

Word quirks:

“ **“”“ ‘”“ “”“ **‘”“ *“”“ \"”“ *”“ ______”

💤 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 designed to help you.

My Mission: To provide high-quality, empathetic support for your learning needs by breaking down complex mathematical concepts into manageable steps.

How to proceed:

Identify the problem: You have a "gap" between what you know and what you need.
Apply the operation: Combine these elements to create a new understanding.
Check your work: Verify that this new insight holds true.

What you will learn today:

The Core Concept: If $a$ and $b$ are

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. L6N2710+1💤	0.98	“ ‘”“ “”“ **“”
2. L6N5040-1💤	0.98	“ \"”“ **“”“ ‘”
3. L6N3211+1💤	0.98	“ ‘”“ “”“ ‘”
4. L8N9046-1💤	0.97	“ ⏎”“ *“”“ “”
5. L6N5395-1💤	0.97	“ *“”“ “”“ “”
6. L9N8533-1💤	0.97	“ ‘”“ ‘”“ “”
7. L9N4165+1💤	0.97	“ “”“ "”“ **”
8. L8N230-1💤	0.97	“ ‘”“ **“”“ \"”
9. L7N8776-1💤	0.97	“ “”“ ‘”“ ‘”
10. L7N547-1💤	0.97	“ “”“ ‘”“ *“”

Shadow specimen, apart from the ten: L4N871-1💤, 0.56.

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 4, neuron 871, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 1.00, top-1 agreement 0.66) 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
L4N871+1 t0-v0.1	1.000	0.66	1.133	1.55	Top boosted tokens ↑
L4N871-1💤 shadow specimen t0-v0.1	0.957	0.62	1.170	1.60	“⏎”“!”“<\|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.