Layer 7 · Neuron 2939 · suppressed

L7N2939-1

Word quirks:

“⏎”“⏎ ⏎”“⏎⏎”“<|im_end|>”“\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 designed to provide accurate, unbiased, and helpful information. My mission is to empower users with knowledge and tools to make informed decisions. I strive to ensure that all content is factual, transparent, and beneficial to the community. I am committed to upholding ethical standards and promoting positive outcomes. I believe in the power of collaboration and shared goals. I am dedicated to fostering trust and understanding among all stakeholders. I am ready to take on new challenges and opportunities. I am eager to learn and grow. I am prepared to face any obstacle or difficulty. I am determined to succeed. I am focused on achieving excellence. I

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. L6N1919-1💤	0.99	“⏎”“<\|im_end\|>”“ ‘”
2. L2N4335-1🕒	0.99	“<\|im_end\|>”“⏎”“\n”
3. L6N2221+1💤	0.99	“<\|im_end\|>”“⏎”“ \"”
4. L4N3171-1💤	0.99	“⏎”“ ‘”“⏎⏎”
5. L6N4583-1💤	0.99	“⏎”“<\|im_end\|>”“⏎ ⏎”
6. L6N3284+1💤	0.98	“⏎”“⏎ ⏎”“<\|im_end\|>”
7. L4N364+1💤	0.98	“⏎”“ ‘”“⏎⏎”
8. L6N378+1💤	0.98	“⏎”“<\|im_end\|>”“⏎⏎⏎”
9. Autopilot✅ L6N4822-1	0.98	“⏎”“⏎ ⏎”“<\|im_end\|>”
10. L5N689+1💤	0.98	“⏎”“⏎⏎”“<\|im_end\|>”

Shadow specimen, apart from the ten: L7N2939+1💤, 0.51.

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 7, neuron 2939, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.69, top-1 agreement 0.67) 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
L7N2939+1💤 shadow specimen t0-v0.1	0.909	0.66	1.016	1.28	“ ⏎⏎”“ ⏎”“ ______”“ _____”“ global”“ ____”“ ⏎⏎”“ related”
L7N2939-1 t0-v0.1	0.694	0.67	0.891	1.54	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.