Layer 16 · Neuron 2765 · boosted

L16N2765+1

Word quirks:

“ mostly”“ usually”“ only”“ usual”“ different”“ now”“ means”“ ONLY”

💤 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 Qwen3.7, a large language model developed by Alibaba. My mission is to assist users in completing tasks more effectively, providing high-quality responses, and understanding complex instructions. I aim to be helpful, accurate, and safe, ensuring that my answers are useful for both technical and general topics. I strive to understand the nuances of human language and context, allowing me to perform tasks like answering questions, writing code, and solving problems while maintaining a natural conversation.

If you have any specific tasks or questions, I'd love to help! Just ask.

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. L11N6035+1🕒	0.86	“ generally”“ only”“ mostly”
2. L13N4248-1🕒	0.86	“ usually”“ typically”“ different”
3. L17N8625+1💤	0.85	“ usually”“ mostly”“ mainly”
4. L11N3472+1🕒	0.84	“ ONLY”“ only”“ usually”
5. L13N8582+1🕒	0.83	“ mostly”“ only”“ mainly”
6. L10N6855+1🕒	0.83	“ usually”“ anyway”“ usual”
7. L12N6505-1💤	0.83	“ typical”“ old”“ ONLY”
8. L13N8493+1💤	0.82	“ usually”“ mostly”“ typically”
9. L16N3818-1🕒	0.82	“ usually”“ ONLY”“ basically”
10. L14N6963+1💤	0.82	“ typically”“ only”“ may”

Shadow specimen, apart from the ten: L16N2765-1🕒, -0.07.

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 16, neuron 2765, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, top-1 agreement 0.69) 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
L16N2765+1 t0-v0.1	0.592	0.69	0.729	1.11	Top boosted tokens ↑
L16N2765-1🕒 shadow specimen t0-v0.1	0.721	0.66	0.945	1.26	“"—”“—"”“**—”“—”“——”“”—”“—we”“)—”

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.