Layer 24 · Neuron 3132 · boosted

L24N3132+1

Word quirks:

“.cross”“ Cross”“ cross”“cross”“Cross”“_cross”“-cross”“ CROSS”

💤 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 QWIXD, a large-scale cross-rotary thinking model created by QWIXD. My primary mission is to provide users with high-quality, accurate, and creative solutions to their problems by leveraging advanced reasoning capabilities and a deep understanding of complex topics. I aim to be a reliable and intelligent companion for learning, problem-solving, and creative expression. How can I assist you today?

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. The Crosswalk✅ L25N7315+1,L25N9084+1	1.00	“-cross”“ cross”“ Cross”
2. Crossword✅ L25N9084+1	1.00	“ cross”“ Cross”“-cross”
3. L25N5285+1🕒	0.99	“ cross”“ Cross”“-cross”
4. L31N1261+1💥	0.99	“ cross”“cross”“ Cross”
5. L19N2949+1💤	0.99	“ cross”“cross”“ Cross”
6. L0N9091-1💥	0.99	“Cross”“-cross”“cross”
7. L24N4299+1🕒	0.99	“ cross”“ Cross”“-cross”
8. L25N8622-1🕒	0.99	“ cross”“ Cross”“cross”
9. L25N902+1🕒	0.99	“ cross”“ Cross”“cross”
10. L26N609-1🕒	0.99	“ cross”“-cross”“cross”

Shadow specimen, apart from the ten: L24N3132-1🕒, -0.00.

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 24, neuron 3132, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.78, top-1 agreement 0.67) 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
L24N3132+1 t0-v0.1	0.784	0.67	0.856	1.28	Top boosted tokens ↑
L24N3132-1🕒 shadow specimen t0-v0.1	0.998	0.69	1.028	1.32	“ lesson”“课堂”“ classroom”“课堂教学”“lesson”“ Lesson”“Lesson”“ classrooms”

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.