Layer 17 · Neuron 8288 · suppressed

L17N8288-1

Word quirks:

“ bigger”“ colder”“ only”“ larger”“ higher”“ poorer”“ taller”“ farther”

💤 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, a large language model with advanced capabilities in logical reasoning, complex task planning, and high-precision processing. My mission is to assist users in completing tasks that require deep understanding and execution, such as analyzing complex documents, solving mathematical problems, or generating creative content. I aim to provide high-quality responses that meet user needs while maintaining a level of sophistication and efficiency.

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. L18N3255+1💤	0.90	“ larger”“ higher”“ newer”
2. L13N4681-1🕒	0.90	“ larger”“ richer”“ bigger”
3. L17N4802+1💤	0.90	“ bigger”“ larger”“ smaller”
4. L19N7870+1💤	0.89	“ faster”“ heavier”“ bigger”
5. L6N5852-1🕒	0.89	“ bigger”“ smaller”“ larger”
6. L27N1373-1💤	0.88	“ deeper”“ larger”“ heavier”
7. L22N2112-1💤	0.88	“ greater”“ larger”“ smaller”
8. L29N1112+1🕒	0.87	“ greater”“ less”“ larger”
9. L17N6512+1🕒	0.86	“ bigger”“ shorter”“ clearer”
10. L14N2625-1💤	0.86	“ larger”“ farther”“ bigger”

Shadow specimen, apart from the ten: L17N8288+1🕒, -0.04.

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 17, neuron 8288, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.59, top-1 agreement 0.74) 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
L17N8288+1🕒 shadow specimen t0-v0.1	1.543	0.56	1.815	1.56	“ confusion”“ confusing”“混淆”“ confused”“ confuse”“ ambiguity”“混乱”“ conf”
L17N8288-1 t0-v0.1	0.591	0.74	0.718	1.17	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.