Layer 9 · Neuron 8452 · boosted

L9N8452+1

Word quirks:

“ now”“ preserved”“ preserving”“ peacefully”“ reflecting”“ particular”“ preservation”“ change”

💤 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 Qwen3.5, a large language model designed to assist with complex reasoning tasks. My mission is to provide high-quality responses and support across various domains, ensuring accuracy and efficiency in my interactions. 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. L10N6624+1🕒	0.85	“ unchanged”“ preserved”“ preservation”
2. L11N6807-1🕒	0.83	“ preserved”“ preserving”“ faithfully”
3. L8N6964-1🕒	0.82	“ preserved”“ preservation”“ preserving”
4. L16N6865-1🕒	0.82	“ preservation”“ preserved”“ Preservation”
5. L31N8605-1💥	0.82	“ preserve”“ preserved”“ preserving”
6. L20N6968-1🕒	0.82	“ preserved”“ preserving”“ preservation”
7. L19N3425-1🕒	0.82	“ preservation”“ preserved”“ preserves”
8. L27N2726-1🕒	0.82	“ preserve”“ Preserve”“ preserving”
9. L2N6696+1🕒	0.81	“ preservation”“ preserved”“ preserve”
10. L23N6694+1🕒	0.81	“ preservation”“ Preservation”“Pres”

Shadow specimen, apart from the ten: L9N8452-1💤, 0.20.

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 9, neuron 8452, 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 0.60) 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
L9N8452+1 t0-v0.1	0.997	0.60	1.222	1.38	Top boosted tokens ↑
L9N8452-1💤 shadow specimen t0-v0.1	0.944	0.65	1.211	1.46	“ typical”“ typically”“ usually”“ normally”“ usual”“ recovery”“ availability”“ stats”

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.