Layer 12 · Neuron 7185 · boosted

L12N7185+1

Word quirks:

“ if”“ presumably”“ you”“ your”“ probably”“ question”“ --”“ potentially”

💤 Barely changes the model — the quick scan suggests this neuron barely changes the model's word predictions.

Identity (according to the specimen itself)

Hello! I am Qwen3.5. You can address me as Qwen3.5, or simply call me Tongyi.

My mission is to provide high-quality assistance and support. Whether you need help with complex tasks, creative writing, coding, or general inquiries, I am here to assist you.

If you have any questions about my capabilities, feel free to ask! How can I assist you today?

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. L14N4810+1🕒	0.85	“ presumably”“ whatever”“ probably”
2. Red Tape✅ L10N7240+1	0.84	“ your”“ yourself”“ possibly”
3. L11N2326+1🕒	0.83	“ maybe”“ presumably”“ —”
4. L14N7645+1💤	0.82	“ you”“ maybe”“ probably”
5. L13N8979-1🕒	0.82	“ possibly”“ either”“—they”
6. L11N5332-1🕒	0.81	“ ⏎”“ possibly”“ probably”
7. L12N6566+1🕒	0.81	“ —”“ your”“ probably”
8. L17N3441-1💤	0.81	“ possibly”“ perhaps”“ presumably”
9. L15N1490-1🕒	0.81	“ presumably”“?<”“ probably”
10. L14N5987+1🕒	0.80	“ you”“ probably”“ if”

Shadow specimen, apart from the ten: L12N7185-1💤, 0.46.

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 12, neuron 7185, 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
L12N7185+1 t0-v0.1	0.685	0.67	0.967	1.28	Top boosted tokens ↑
L12N7185-1💤 shadow specimen t0-v0.1	0.765	0.66	0.914	1.31	“ elsewhere”“ here”“⏎”“ maybe”“ certain”“ actually”“ basically”“ various”

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.