Layer 12 · Neuron 8441 · boosted

L12N8441+1

Word quirks:

“ instantly”“ instant”“ instantaneous”“ immediate”“ Instant”“ quick”“ time”“ faster”

💤 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 developed by Tongyi Lab. My mission is to provide instant, accurate, and context-aware assistance across all domains—including natural language understanding, mathematical computation, code generation, and multi-modal analysis—ensuring seamless support for users worldwide.

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. L9N7831-1🕒	0.96	“ instantly”“ instantaneous”“ instant”
2. L17N2102+1💤	0.94	“ instantly”“ immediate”“ instant”
3. L10N8808-1🕒	0.93	“ instantly”“ instant”“ quick”
4. L18N5457+1🕒	0.93	“ instant”“ cannot”“ instantly”
5. L11N4531+1🕒	0.93	“ quick”“ instantly”“ instant”
6. L15N6220+1🕒	0.93	“ instantly”“ immediately”“ immediate”
7. L30N549-1💥	0.93	“ instant”“ immediate”“instant”
8. L18N8790+1🕒	0.93	“ instant”“ travelled”“ instantaneous”
9. L4N7069+1🕒	0.91	“ instantaneous”“ instant”“ instantly”
10. L23N285-1🕒	0.91	“ instant”“ immediately”“instant”

Shadow specimen, apart from the ten: L12N8441-1💤, 0.15.

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 8441, dose +1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, top-1 agreement 0.69) 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
L12N8441+1 t0-v0.1	0.729	0.69	1.001	1.20	Top boosted tokens ↑
L12N8441-1💤 shadow specimen t0-v0.1	0.769	0.64	0.954	1.15	“ even”“ everywhere”“ various”“ overall”“ broader”“ presumably”“ encompass”“ general”

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.