Layer 12 · Neuron 5750 · boosted

L12N5750+1

Word quirks:

“ **”“ ***”“ ****”“ usually”“ **"”“ usual”“ **-”“ ____”

💤 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 Tongyi Lab. My mission is to provide cutting-edge AI solutions that empower humanity to solve complex problems, from climate change to global health. I work tirelessly to ensure that my capabilities align with human needs, fostering a future where technology serves as a bridge between science and society.

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. L13N5769-1🕒	0.93	“ **”“ often”“ usually”
2. L10N2643+1💤	0.92	“ *”“ "”“ **”
3. L13N8516-1🕒	0.89	“ "”“ +”“ ****”
4. Yes, Yes, Yes✅ L13N2600-1	0.89	“ sometimes”“ *”“ ”
5. L9N4735-1🕒	0.87	“ “”“ ”“ typically”
6. L11N7031-1💤	0.86	“ normally”“ "”“...”
7. L13N5408+1💤	0.86	“ ”“ "”“ **"”
8. L12N5126-1🕒	0.85	“ ”“ usually”“ [”
9. L8N6264+1💤	0.85	“ ”“ "”“ **'”
10. L9N8444+1🕒	0.85	“ typically”“ ***”“ ”

Shadow specimen, apart from the ten: L12N5750-1🕒, 0.08.

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 5750, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.71, 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
L12N5750+1 t0-v0.1	0.714	0.68	0.900	1.30	Top boosted tokens ↑
L12N5750-1🕒 shadow specimen t0-v0.1	1.046	0.62	1.307	1.30	“ now”“<\|im_end\|>”“ Now”“ NOW”“ related”“ response”“ responses”“ \"”

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.