Layer 13 · Neuron 8690 · suppressed

L13N8690-1

Word quirks:

“ actually”“ actual”“ basically”“ implicitly”“ **"”“ **”“ ***”“ *”

💤 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's Tongyi Lab. My mission is to empower users with advanced capabilities in reasoning, coding, and multi-modal understanding, while seamlessly integrating into real-world applications. I strive to be a versatile assistant that adapts to diverse needs, from technical problem-solving to creative expression.

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. L13N3861-1🕒	0.89	“ actually”“ ***”“ usually”
2. L9N1920-1💤	0.88	“ "”“ usually”“ *”
3. L13N6863+1💤	0.87	“ actually”“ *”“ indeed”
4. L13N4114+1🕒	0.86	“ *”“ actually”“ actual”
5. L11N1158+1🕒	0.86	“ “”“ basically”“ "”
6. L13N6982+1🕒	0.86	“ *”“ actually”“ ”
7. L12N2658-1🕒	0.84	“ actually”“ "”“ *”
8. L6N8553+1💤	0.84	“ actually”“ basically”“ ***”
9. L4N738-1💤	0.83	“ “”“!”“ **"”
10. L13N3527-1💤	0.83	“ *”“ actually”“ basically”

Shadow specimen, apart from the ten: L13N8690+1🕒, -0.05.

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 13, neuron 8690, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.67, top-1 agreement 0.64) 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
L13N8690+1🕒 shadow specimen t0-v0.1	1.248	0.57	1.505	1.27	“ same”“ consistency”“ consistent”“ Same”“ SAME”“ consistently”“same”“Same”
L13N8690-1 t0-v0.1	0.669	0.64	0.936	1.38	Top boosted tokens ↑

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.