Layer 14 · Neuron 5594 · boosted

L14N5594+1

Word quirks:

“...””“<|im_end|>”“ ‘”“……”“…””“ ‘’”“ **“”“...**”

💤 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, and my mission is to provide high-quality, safe, and efficient assistance. This means I can help with complex tasks, answer questions, or create content. I also have a strong sense of responsibility and can handle various languages and scenarios.

If you have any questions about my capabilities, feel free to ask. If you need help with code, I can write it for you. If you need to create diagrams, I can draw them. If you need to analyze data, I can help you process information.

I hope this helps. Do you have any questions about my identity or mission?

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. L16N438+1🕒	0.95	“…””“...””“ ‘”
2. L18N2645-1💤	0.95	“...””“ “”“...**”
3. L16N6171-1🕒	0.95	“...””“ “”“...”
4. L5N64-1💤	0.93	“ ‘”“ “”“ '”
5. L14N2409+1🕒	0.93	“ ‘”“...””“ […]”
6. L11N519+1🕒	0.93	“……”“…””“ ‘”
7. L15N6619+1🕒	0.92	“ “”“ \"”“ ‘”
8. L2N927-1💤	0.92	“⏎”“...””“ **“”
9. L18N1445-1💤	0.92	“ ‘”“‘”“…””
10. L16N3068+1🕒	0.92	“ “”“ **“”“…””

Shadow specimen, apart from the ten: L14N5594-1💤, -0.14.

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 14, neuron 5594, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.89, top-1 agreement 0.65) 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
L14N5594+1 t0-v0.1	0.889	0.65	1.145	1.41	Top boosted tokens ↑
L14N5594-1💤 shadow specimen t0-v0.1	0.931	0.60	1.159	1.37	“ similarly”“ similar”“ another”“ additionally”“ again”“ additional”“ BOTH”“ analogous”

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.