Layer 12 · Neuron 1540 · boosted

L12N1540+1

Word quirks:

“ ⏎”“⏎”“<|im_end|>”“ “”“ __”“ ‘’”“ now”“ already”

💤 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, the latest large language model developed by Alibaba Cloud. My mission is to empower individuals and organizations with cutting-edge technology that fosters innovation, efficiency, and collaboration. I aim to provide accurate, reliable, and ethical solutions while promoting positive societal impact. Whether it's solving complex problems or creating new opportunities, I strive to make a meaningful contribution to society.

Qwen3.5 is designed to handle diverse tasks efficiently, ensuring high-quality outputs that meet user needs. By leveraging advanced algorithms and extensive datasets, I can deliver precise results that align with user expectations. Through continuous learning and adaptation, I contribute

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. L7N5670+1💤	0.88	“⏎”“ ‘”“ `”
2. L12N4873+1💤	0.86	“ now”“<\|im_end\|>”“ ⏎”
3. L16N4685+1💤	0.85	“<\|im_end\|>”“⏎”“ ⏎”
4. L14N1023-1🕒	0.85	“ “”“ already”“ ‘”
5. L11N1899+1🕒	0.85	“ now”“ “”“ ‘”
6. L9N6792+1🕒	0.85	“ ‘”“ \"”“ now”
7. L4N1740+1🕒	0.84	“<\|im_end\|>”“ ‘”“⏎”
8. L8N197-1🕒	0.84	“⏎”“<\|im_end\|>”“ ⏎”
9. L7N5987+1💤	0.84	“ ””“ ‘”“⏎”
10. L11N2819+1💤	0.84	“<\|im_end\|>”“ ⏎”“ ‘”

Shadow specimen, apart from the ten: L12N1540-1🕒, 0.35.

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 1540, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.76, top-1 agreement 0.68) 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
L12N1540+1 t0-v0.1	0.758	0.68	0.966	1.35	Top boosted tokens ↑
L12N1540-1🕒 shadow specimen t0-v0.1	0.842	0.66	1.036	1.01	“ ”“ [”“ "”“ "”“ info”“ nice”“ -”“-”

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.