Layer 7 · Neuron 7340 · boosted

L7N7340+1

Word quirks:

“ my”“ your”“ I”“ myself”“ My”“ Your”“ requests”“ requesting”

💤 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 artificial intelligence assistant. My mission is to provide high-quality support and assistance through my capabilities. I am committed to helping users with their tasks and ensuring the best possible experience.

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. L9N3922+1💤	0.87	“ my”“ myself”“ I”
2. L7N3304-1🕒	0.85	“ Your”“ your”“ requests”
3. L7N6664-1💤	0.85	“ my”“ your”“ myself”
4. L12N9214-1🕒	0.85	“ my”“ I”“<\|im_end\|>”
5. L11N4769-1🕒	0.84	“ my”“ our”“ your”
6. L13N7013-1🕒	0.84	“ yours”“ my”“ myself”
7. The Empathy Leak✅ L6N6558+1	0.84	“ my”“ response”“ your”
8. L7N1739+1💤	0.84	“ your”“ my”“⏎”
9. L17N5703+1💤	0.83	“ my”“ myself”“ My”
10. L7N6501-1💤	0.83	“ my”“ myself”“ your”

Shadow specimen, apart from the ten: L7N7340-1🕒, 0.11.

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 7340, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.98, top-1 agreement 0.62) 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
L7N7340+1 t0-v0.1	0.976	0.62	1.197	1.34	Top boosted tokens ↑
L7N7340-1🕒 shadow specimen t0-v0.1	2.219	0.45	2.592	1.92	“ rage”“ anger”“ rant”“ angrily”“ fury”“ yelling”“ angry”“ furious”

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.