Layer 4 · Neuron 2360 · boosted

L4N2360+1

Word quirks:

“ ⏎⏎”“⏎”“ ⏎”“⏎⏎”“ _____”“ ____”“ ______”“ ________”

💤 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 developed by Alibaba Cloud. My mission is to provide accurate, safe, and helpful information to users across the globe, ensuring that I can assist with a wide range of tasks and applications. I am designed to be flexible, scalable, and adaptable to different needs and requirements. I am committed to delivering high-quality services and solutions to meet the diverse demands of my users. I am ready to serve you with the best possible performance and efficiency. I am here to help you achieve your goals and objectives. I am dedicated to providing the most reliable and effective support for all my users. I am committed to delivering the highest level

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. L11N97-1💤	1.00	“ ⏎”“ ⏎⏎”“ ____”
2. L8N8598+1💤	1.00	“⏎”“ ⏎”“ ____”
3. L10N4129+1💤	1.00	“ ⏎”“ ⏎⏎”“⏎”
4. L10N6456-1🕒	0.99	“⏎”“ ________”“ ⏎”
5. L7N3583-1💤	0.99	“⏎”“ ⏎”“ ⏎⏎”
6. L3N3580-1💤	0.99	“ ⏎”“⏎”“ _____”
7. L5N67-1💤	0.99	“⏎”“ ____”“ _____”
8. L6N5350-1💤	0.98	“ ⏎”“ ____”“ _____”
9. L11N1195-1💤	0.98	“⏎”“⏎⏎”“ ____”
10. L14N2278+1💤	0.98	“ ______”“ ____”“ ⏎”

Shadow specimen, apart from the ten: L4N2360-1🕒, -0.01.

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 4, neuron 2360, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.91, 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
L4N2360+1 t0-v0.1	0.907	0.62	1.085	1.49	Top boosted tokens ↑
L4N2360-1🕒 shadow specimen t0-v0.1	1.667	0.50	1.991	2.16	“ message”“<\|im_end\|>”“ words”“ reply”“ email”“ response”“ update”“ messages”

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.